Re: Unable to get soundcard working
On Mon, Feb 17, 2003, Tzafrir Cohen wrote about "Re: Unable to get soundcard working": > no. /etc/modules . Something simple I can't find in redhat and Mandrake, > for some reason: a list of modules loaded at startup . Not difficult to > create, but why do it yourself? > > Don't confuse this with /etc/modules.conf which is the config file of > insmod and modprobe Actually, /etc/modules.conf is typically use to load your sound-card module, but in a round-about way: using aliases to char-major-*. For example, /dev/dsp is a character-special device with major number 14. So I have in /etc/modules.conf something like alias char-major-14 soundcore This means the soundcard module will only get loaded the first time it is needed. This soundcore stuff is for ALSA; Try redhat-config-soundcard for something that will hopefully work for your card. In modules.conf you can also specify options to modules, specify programs to run when the module is loaded/unload (e.g., I use it to save the volume setting across module reload and even reboots). -- Nadav Har'El| Monday, Feb 17 2003, 15 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |When you handle yourself, use your head; http://nadav.harel.org.il |when you handle others, use your heart. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Guessing filesystem while unmounted
On Mon, Feb 17, 2003, Muli Ben-Yehuda wrote about "Re: Guessing filesystem while unmounted": > You can do it the same way the kernel does it in the early boot stages > - by getting the information from lilo, via the boot record. You'll > need to read and parse the boot record, though. Mounting each file > system and checking may very well be easiser... If this is supposed to be a "hack" and not something that must generally work, you can try "fdisk -l" and look for bootable Linux partitions. In many installations the root partition will be exactly of this type. On other installations, though, you might catch a special "/boot" partition, not the "/" partition, or none at all (lilo doesn't strictly need the partition to be bootable). Note, by the way, that newer distributions (such as Redhat 8) use Grub, not LILO, so anything relying on the structure of a lilo boot record isn't foolproof either. -- Nadav Har'El| Monday, Feb 17 2003, 15 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The world is coming to an end ... SAVE http://nadav.harel.org.il |YOUR BUFFERS!!! = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Guessing filesystem while unmounted
On Mon, Feb 17, 2003, Muli Ben-Yehuda wrote about "Re: Guessing filesystem while
unmounted":
> On Mon, Feb 17, 2003 at 05:48:07PM +0200, Nadav Har'El wrote:
> > Note, by the way, that newer distributions (such as Redhat 8) use Grub,
> > not LILO, so anything relying on the structure of a lilo boot record
> > isn't foolproof either.
>
> I never investigated this, so take it with a grain of salt, but I
> believe that the boot record structure is standardized and even *gasp*
> documented.
Well, as far as I know Grub does not know where your root filesystem ("/")
lives, until it first finds the boot filesystem, finds the grub/grub.conf
file in it (grub understands the ext2 filesystem), reads it and finds there
the specification of where the filesystem is.
This is both quite complex, and very different from what Lilo does; With
lilo, you must rerun "lilo" (to change the boot sector) every time you
make a change to /etc/lilo.conf. With grub, you don't have to do it, because
most of the information is looked up on your boot filesystem, not on the
boot sector.
--
Nadav Har'El| Monday, Feb 17 2003, 16 Adar I 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Sign seen in restaurant: We Reserve The
http://nadav.harel.org.il |Right To Serve Refuse To Anyone!
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Guessing filesystem while unmounted
On Mon, Feb 17, 2003, Orna Agmon wrote about "Re: Guessing filesystem while unmounted":
> > Well, as far as I know Grub does not know where your root filesystem ("/")
> > lives, until it first finds the boot filesystem, finds the grub/grub.conf
> > file in it (grub understands the ext2 filesystem), reads it and finds there
> > the specification of where the filesystem is.
>
> What happens if you use another filesystem on / ? ext3, for example. Or
> something that is not ext3 at all? can grub read all of them?
Ext3 is basically backward-compatible with Ext2, in the sense that you can
mount a clean Ext3 system as Ext2 without losing anything. So Grub didn't
need to make any changes to read Ext3 filesystems
Anyway, "info grub" is your friend when it comes to grub. Quoting from that
info,
"The currently supported filesystem types are "BSD FFS", "DOS FAT16
and FAT32", "Minix fs", "Linux ext2fs", "ReiserFS", "JFS", "XFS",
and "VSTa fs".
If you look on your disk, in directory /boot/grub, you'll notice files
like "e2fs_stage1_5", "reiserfs_stage1_5", etc. - these are (as far as I
know - I'm not really a grub expert...) the files that grub uses to be
able to later read the configuration file, stage2, and finally the actual
kernel, from the file system. Some of this is explained in the info file
(check "hacking GRUB").
--
Nadav Har'El| Monday, Feb 17 2003, 16 Adar I 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Boat: A hole in the water surrounded by
http://nadav.harel.org.il |wood into which one pours money.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: ISOC day at Kfar HaMakabiya
On Tue, Feb 18, 2003, Reuven M. Lerner wrote about "ISOC day at Kfar HaMakabiya": > Ira> anyone been there and cares to share any news and views they > Ira> collected? how was GBY's talk about Hamakor? >... > (c) he was too Stallman-like for my taste, Too Stallman-like? Gilad, you can take that as a compliment :) > and (d) he didn't talk very > much (if at all) about Hamakor -- but that might have been because of > time constraints. Those of you who attended the Linux day at IBM probably remember someone who was invited to talk about some subject, and could not shut up about his own company, attributing half of the development of computers in the last two decades to his (very young) company. So it's probably for the best that Gilad can talk about the goals and motives of his organization, not only the organization itself, which with all due respect, hasn't really done anything yet. (but note that I did not attend this conference, so I didn't actually hear him). > conference, it sounds like things are moving ahead steadily on > OpenOffice.org, but that there's still a wide gap between what the > average user needs and what will be available shortly. Nevertheless, Can you give a few more details? What generally-useful features of OpenOffice won't be available in Hebrew initially? (shameless plug) at least somebody is working on a spellchecker :) -- Nadav Har'El| Tuesday, Feb 18 2003, 16 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I used to be a lumberjack, but I just http://nadav.harel.org.il |couldn't hack it, so they gave me the axe. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Guessing filesystem while unmounted
On Tue, Feb 18, 2003, Uri Itscowits wrote about "Re: Guessing filesystem while
unmounted":
> First I want to thank everybody for the time & effort,
> ( although some responses show,
> some guys did NOT spend enough time reading my problem all the way {:< )
:)
I hope my response wasn't one of those. I did read your message, but frankly
it wasn't very easy to understand what you're trying to do. In particular it
isn't clear if you are looking for a ad-hoc hack that will work on a particular
system, or some universal set of heuristics that will work on old systems
(without e2labels, with lilo, bootable Linux partition, etc.) as well as new
(with e2labels, grub, etc.).
It also wasn't clear why you are so against trying to mount the partitions one
by one (the list of partitions is in "fdisk -l", as I said) and checking if
they have "/-like" files (/bin/ls, /root, /etc/passwd, etc.).
> Since no one gave me a good answer,
> (except for Muli's, which was my idea to begin with)
Be careful, if you are invested in an answer before you ask the question, you
might not be open to better, or at least different, answers ;)
> PS
> no need to CC me in person, I'm on the list, & it floods my mailbox.
Tough luck. Join the pro-Reply-To: crowd (of which I'm a proud member).
--
Nadav Har'El| Tuesday, Feb 18 2003, 16 Adar I 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Microchips: what's left at the bottom of
http://nadav.harel.org.il |the bag when it reaches you.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Guessing filesystem while unmounted
On Tue, Feb 18, 2003, Ira Abramov wrote about "Re: Guessing filesystem while unmounted": > Quoting [EMAIL PROTECTED], from the post of Tue, 18 Feb: > > That was unintentional and might go to show my point that I couldn't > > tell (or it was very obscure to me) whether your message was private > > or public. > > if your MUA is so "dumbed down" that you can't tell if a mail was sent > to you, the list or both, it means it's hiding the recipiænt headers, so > what's left of the message? It means that when somebody knows the "r"eply doesn't work on linux-il and is always used to add the CC: back to the list himself, he can accidentally do that even on messages that just appear to have come from the list (because of the people involved) but infact haven't. I'm not saying that's what happened in this case, but it's a possibility. I use "group reply" instead of an explicit CC:, but not everyone does! See also your (Ira) own complaints that people take messages he sent to iglu.org.il and respond to them in linux.org.il. See - when people complain that Reply-To: causes embarrassing mistakes, they forget that such things also happen without it. This guy (may have) accidentally replied on the list. I once got legal threats for having used "group reply" (the jobinfo incident). Now, let's drop this issue - it has barely been a month since we last argued it :( And Shachar - this time it's not my fault we're arguing this :) -- Nadav Har'El| Tuesday, Feb 18 2003, 16 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |What did the Buddhist say to the hot dog http://nadav.harel.org.il |vendor? Make me one with everything. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Bork, Bork, Bork!
On Sun, Feb 16, 2003, Nadav Har'El wrote about "Bork, Bork, Bork!": > According to > http://www.ynet.co.il/articles/0,7340,L-2439013,00.html > > Opera has a new special version which "messes" with MSN.com, translating its > text into "swedish chef" language [1]. By the way, for Swedish Chef fans, Opera supporters, and Burger eaters, you might be interested to know that McDonald's Happy Meals now contain small Muppets dolls, among them our friend, Bork Bork Bork. You can also get it when you buy a normal combo, if you add 4.90 shekels. Note that they don't always have all the dolls in stock, so unless you want a Kermit to protest against Microsoft's Hyperterminal, check before you order :) -- Nadav Har'El| Tuesday, Feb 18 2003, 16 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Today is the tomorrow you worried about http://nadav.harel.org.il |yesterday, and now you know why. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: need tool for high quality typesetting, unicode-capable
On Wed, Feb 19, 2003, [EMAIL PROTECTED] wrote about "need tool for high quality typesetting, unicode-capable": > Hi, > > I need to design a Chumash with Hebrew and German text, and would like it to be You might want to take a look at Makor (http://www.ctan.org/tex-archive/language/hebrew/makor/). It matches most of your requirements, but I'm not sure all of them. Note that I never used it - I only read the README - so I can't comment on its actual functionality. Check out its README (link about), manual at http://www.ctan.org/tex-archive/language/hebrew/makor/tex/makor2/mkr2man.pdf (including some impressive examples!) Or http://www.dsm.fordham.edu/~agw/makor-sample.ps for a small example. (for some reason, the PDF versions have quite a few distorted letters - I don't know why; as I said I never used makor myself). -- Nadav Har'El| Wednesday, Feb 19 2003, 17 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A computer program does what you tell it http://nadav.harel.org.il |to do, not what you want it to do. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: need tool for high quality typesetting, unicode-capable
On Wed, Feb 19, 2003, Ira Abramov wrote about "Re: need tool for high quality typesetting, unicode-capable": > http://www.index.co.il/dfus/article_page.asp?info_id=112840 >... very interesting links! > Sadly, the Bible is partly a project of the Hebrew Uni, but I see no > publication of theirs on the subject. If there was any justice, all the > information from and about the project would have been available to the > public, including fonts and sources. Right, the hiding of information in Israeli universities and other state- funded institutions is a real shame. American univerisities produced BSD (UC Berkeley), X11 (MIT), Tcl/Tk (UC Berkeley), Kermit (Columbia), and tons of other stuff, while Israeli universities produced - what? And when finally some good free software does come out of an Israeli university (Mosix), people start to complain about it and the only righteous man in Sodom that wrote it, and fork it off. As Seinfeld likes to say, "That's a shame...". Luckily, if you want to get the text of the bible, it is avaliable, in http://www.mechon-mamre.org/. That is actually a very good site: you can get the entire bible in various formats (ktiv male, with niqqud, with te`amim, etc.), as well as browse or search it online. I use it for all my biblical needs :) However, I don't know how "official" or "verified" their version of the texts is. I also don't know what their license for the text is - though in my opinion (and I'm not a lawyer) Israeli copyright laws recognize only innovation, not hard work, so a online bible that basically contains the same text as existing bible *can not* be copyrighted. (If anybody knows this to be false, please correct me). P.S. The niqqud-less versions of their site looks beautiful in Mozilla. Unfortunately, the ones with niqqud look horrible, because Mozilla renders those on the side, not below the letters. Does anybody on this list knows when, if ever, Mozilla are planning to support Hebrew diacritical marks? I'm also interested in Arabic diacritics, that also don't work properly :( -- Nadav Har'El| Wednesday, Feb 19 2003, 18 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Despite the cost of living, have you http://nadav.harel.org.il |noticed how it remains so popular? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: need tool for high quality typesetting, unicode-capable
On Wed, Feb 19, 2003, Shlomi Fish wrote about "Re: need tool for high quality typesetting, unicode-capable": > Of course, I don't know how many of these projects are of actual > substantial worth. Some labs give exactly the same project semester after They are usually of near-zero substantial worth unfortunately. Just like the homework I did while studying math was rarely (if ever!) worth publication. It was usually just repeating what was already done N+1 times in the past, and many times - badly. I was talking about major projects taking several experienced people (like professors and graduate students) and spanning years, not about something a single completely inexperienced person does in two days (or even two weeks) of effort. > In any case, the reason openMosix was forked (in part by Israelis) was > because Prof. Amnon Barak does not accept patches to it from the outside. This is a valid reason for forking, exactly like it is valid for me to declare "NYH Linux" based on Linux with a few patches I wrote and which Linus refused to enter into the official kernel. What is not ok, however, is to go around saying that my version is "Open Linux" because Linus's is closed. It's simply not fair. Especially when you do it to one of the "good guys" (like Prof. Barak or Linus Torvalds). > That said, I believe that software that is developed inside universities > and not released to the public is not a good or advisable thing. It is > possible that it technically legal. In the U.S. there's the issue of the > stanford checker, which was used to find some bugs in the Linux kernel, > but has otherwise not been made available in source or binary forms. The tradition in Universities has always been to publish their results, *and* give enough information in that publication for the readers to be able to replicate the work, and build on it. Does it sound like free software, where you have to publish both the binary *and* the source code needed to replicate this work and build on it? It sure does! But unfortunately some researchers have a different take on this. They think that contrary to the text of the GPL, nobody ever said that the information a university researcher publishes has to be the "most convenient" way to replicate their work. So they don't publish source-code, but only a brief description of what they done. Their "competitors" (other researchers in the same field) can repeat their work, but it will take them months of work, rather than 5 minutes to get started. The researchers that do that probably think that they are still following the university procedures while giving themselves a competitive edge, and they are right, actually. But I think while they are following the procedures, they are not following the spirit for which Universities we designed. You might compare this situation to university departments that have expensive tangible property, like expensive physics measurement devices, an expensive super-computer, a very good library of ancient scrolls, etc. Such "property" gives this university a competitive edge, attracting good researchers and more funds to it. Having such property is good for the University, so the regents of the university probably like it. For some departments, software can actually play the role of such property, so perhaps it's understandable why Universities don't really mind (to say the least) when their researchers keep software hidden. And it certainly doesn't help that certain American univerisities got stakes in high-profile software patents (e.g., see Akamai), so they become greedy and try to capitalize on the research their people are doing, instead of telling them to go ahead and share their work with the world. -- Nadav Har'El| Wednesday, Feb 19 2003, 18 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |What did the Buddhist say to the hot dog http://nadav.harel.org.il |vendor? Make me one with everything. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Mazal Tov! (fonts)
On Thu, Feb 20, 2003, Ira Abramov wrote about "Re: Mazal Tov! (fonts)": > Quoting Hetz Ben-Hamo, from the post of Thu, 20 Feb: > > I have just finished installing Red Hat 8.0.94 and what-do-you-know - > > some well known fonts are in (as long as you select Hebrew when you're > > installing Red Hat).. > > > > Font names: Aharoni, David, Frank Rehul, Miriam mono (fixed), and > > how so? these are hardly free... RHL giving the finger to Microsoft is > one thing, but the copyright holders of these fonts are Israeli font > designers AFAIK, nd they are not freely redistributable... I assume Hetz is talking about the Kulmus fonts, which are free. If it's true, it's really good news :) Maybe I can finally remove the last relic I have of Microsoft on my hard-disk - their fonts! -- Nadav Har'El|Thursday, Feb 20 2003, 18 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I used to be a lumberjack, but I just http://nadav.harel.org.il |couldn't hack it, so they gave me the axe. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: MORE [was: Equipment - anyone?]
On Fri, Feb 21, 2003, Official Flamer/Cabal NON-Leader wrote about "MORE [was: Equipment - anyone?]": > All stuff has been taken ;-) Where do you get all that weird stuff? :) Can I trade you on Commodore 64, one VIC 20, and one 3B1 for a Microvax? (just kidding ;)). Anyway, here's an idea for you - auction off your hardware in exchange to promisses to work on free software. -- Nadav Har'El|Saturday, Feb 22 2003, 20 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Having a smoking section in a restaurant http://nadav.harel.org.il |is like having a peeing section in a pool = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: MORE [was: Equipment - anyone?]
On Sat, Feb 22, 2003, Dan Armak wrote about "Re: MORE [was: Equipment - anyone?]": > This is probably getting OT, but I'll take this opportunity to ask: anyone > know where I can get a low-end usb sound card? > There are several on the market in the 40-50 $ range (fex. Xitel HiFi Link, > Griffin iMic, etc) but I haven't been able to find a single one in Israel, > only the really costly ones like the Extigy, which have lots of features I > can't use anyway. I bought my soundcard in Universe Club (yes, that discount supermarket). Cost me 60 shekels (!) two years ago. You can also go to your favorite dirt-cheap store, and look for the cheapest sound card they carry. Then go home and verify that it is supported in Linux. I did have some trouble with mine initially: I never checked in advance if Linux supports it and, lo and behold, it didn't (two years ago). A commercial driver was available, but it cost 30$ (twice the price of the card!), so obviously I wouldn't buy it - and used the "demo version" for a few months - until ALSA (the advanced linux sound architechture) people started supporting this card - and ever since I'm very happy with this card: I listen to music hours every day with this card. > - - old 386s and P1s and parts of all kinds - old EDO modules, lots of ancient > sound cards, cdroms... I suppose noone wants that kind of stuff because we > all have our own doorstops already :-) Yes, but do you already have a custom-built mp3 player? :) Old computers can become interesting DIY projects! -- Nadav Har'El|Saturday, Feb 22 2003, 20 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A man is incomplete until he is married. http://nadav.harel.org.il |After that, he is finished. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: MORE [was: Equipment - anyone?]
On Sat, Feb 22, 2003, Nadav Har'El wrote about "Re: MORE [was: Equipment - anyone?]": > On Sat, Feb 22, 2003, Dan Armak wrote about "Re: MORE [was: Equipment - anyone?]": > > This is probably getting OT, but I'll take this opportunity to ask: anyone > > know where I can get a low-end usb sound card? > > I bought my soundcard in Universe Club (yes, that discount supermarket). > Cost me 60 shekels (!) two years ago. Oops, I'm an idiot... I didn't see your key request: USB. The card I mentioned is PCI... Sorry. -- Nadav Har'El|Saturday, Feb 22 2003, 20 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |In God we Trust -- all others must submit http://nadav.harel.org.il |an X.509 certificate. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: need tool for high quality typesetting, unicode-capable
On Fri, Feb 21, 2003, Matitiahu Allouche wrote about "Re: need tool for high quality typesetting, unicode-capable": > Nadav Har'El wrote: > P.S. The niqqud-less versions of their site looks beautiful in > Mozilla. > Unfortunately, the ones with niqqud look horrible, because Mozilla renders > those on the side, not below the letters. Does anybody on this list knows >.. > As far as I remember, the problem is not entirely (or maybe at all) with > the Mozilla display engine but with its interaction with fonts. The > solution is to try various fonts until you find one which gives acceptable > results. I am now writing offline, but I seem to remember that Arial > displays Hebrew vowels correctly, except for the Holam which is too far > from its base letter (according to my taste). I found one font (from those > distributed with Windows) which displayed all vowels nicely, but I don't > remember which :-( Can anybody else confirm this? Was anybody able to view Niqqud correctly on Mozilla? For an example, try it on Genesis 1: http://www.mechon-mamre.org/i/t/t0101.htm How do I configure my fonts (and which fonts should I used) for this to work? Does it work with Culmus fonts? What will it take to correct the fonts and/or Mozilla for this to work for all fonts? -- Nadav Har'El|Saturday, Feb 22 2003, 21 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A conscience does not prevent sin. It http://nadav.harel.org.il |only prevents you from enjoying it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Mystifying makefile error
On Sat, Feb 22, 2003, Erez Boym wrote about "Mystifying makefile error": > Hi > > Running Make I get an error which I am unable to > resolve, the error is: > Makefile:164: *** missing separator. Stop. > > line : > @AMDEP_TRUE@@am__include@ > @[EMAIL PROTECTED](DEPDIR)/[EMAIL PROTECTED]@ What the heck is that? I think this is some sort of "automake" input file, which you forgot to run "automake". It certainly doesn't have a "makefile" syntax, so no wonder "make" doesn't like it... -- Nadav Har'El|Saturday, Feb 22 2003, 21 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Today is the tomorrow you worried about http://nadav.harel.org.il |yesterday, and now you know why. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problem with Pth or make or what?
On Sat, Feb 22, 2003, Oron Peled wrote about "Re: Problem with Pth or make or what?": > Obviously! This also point to a bad habbit: > Daniel you have '.' in your path!!! It's not necessarily a bad habbit... Once upon a time, this was considered good practice for non-root users. But calling your programs "test" is indeed a bad habbit :) I've seen more than one person bit by this. > Another related issue. I hope nobody don't use '.' in your path > as root -- this is suicidal in terms of security. You are scaring the newbies :) Let's make one thing clear: this advice comes from the days of multi-user Unix machines, not of personal Linux machines. In the scary scenario, a superuser might cd to some user's directory (hopefully for some legitimate reason), run "ls", and, lo and behold - the user might have a "ls" program in his own directory formatting the disk (or adding a backdoor, or whatever). On a machine used by a single person, it doesn't matter what your path is. If someone already cracked your machine to insert a program, he could probably do whatever he wants anyway. -- Nadav Har'El| Sunday, Feb 23 2003, 21 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Unlike Microsoft, a restaurant would not http://nadav.harel.org.il |charge me for food I find a bug in! = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problem with Pth or make or what?
On Sun, Feb 23, 2003, Shlomi Fish wrote about "Re: Problem with Pth or make or what?": > > > Another related issue. I hope nobody don't use '.' in your path > > > as root -- this is suicidal in terms of security. > > Only on systems which (might) have malicious users. Not relevant for > > home computers. > It is not entirely unlikely that home computers will be penetrated or > compromised while being connected to the Internet. It is still a bad idea > there. One quite unlikely scenario in which you can indeed benefit from not having '.' in your path: someone broke into an account which doesn't belong to a real user (say, httpd) and is unable to upgrade to superuser, so he puts an "ls" in some directory he can write (e.g., /tmp) and hopes the superuser will accidentally run it. Or perhaps a normal user runs it, and then the trojan can modify his setup (add an 'su' alias or program, etc.) to steal the superuser password. Some people might consider this risk serious enough to change the path. I don't, usually. There are plenty of other, more serious, risks. -- Nadav Har'El| Sunday, Feb 23 2003, 21 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The person who knows how to laugh at http://nadav.harel.org.il |himself will never cease to be amused. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Hebrew with mutt
On Sun, Feb 23, 2003, Eli Segal wrote about "Re: Hebrew with mutt":
> When I use less as describe .. how can i reply or view the attachments ???
When you use an external pager with mutt, you have to finish viewing the
current message first ("q" in less), and then you are returned to mutt,
where you can press the normal commands: "r"eply or "v"iew attachments.
--
Nadav Har'El| Sunday, Feb 23 2003, 21 Adar I 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Make it idiot proof and someone will make
http://nadav.harel.org.il |a better idiot.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problem with apache.
On Sun, Feb 23, 2003, Josh Roden wrote about "Problem with apache.": > I receive the following in the browser window when I tried to submit an > excercise thru our site at hadassah: > > Internal Server Error Indeed looks like an error on the server - there's nothing you (as a user) can do about it. Try to complain to [EMAIL PROTECTED], or your local computer-center attache. -- Nadav Har'El| Sunday, Feb 23 2003, 21 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Topologist, n.: A person who cannot tell http://nadav.harel.org.il |a doughnut from a coffee mug. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: 2nd_MORE [was: Equipment - anyone?]
On Sun, Feb 23, 2003, Alon Weinstein wrote about "RE: 2nd_MORE [was: Equipment - anyone?]": > Since this is becoming a main-stream thing to post the list of junk, here's > what I have: Yes, this list is turning into a regular alte-zachen :) Your junk seems quite useful, actually: > 16 port 10Mbit Hub > 8 port 10Mbit Hub (w/ one additional BNC port) These, assuming they have modern UTP ports, are probably quite useful for home network. Unless you're playing movies over this network, 10Mbps is more than enough. > 4 IDE CD-ROM drives These are useful for the mp3-playing machine I suggested last time :) You could make yourself a jukebox! > I'm looking for an old 486 -- case, motherboard, processor & 16-32mb ram. no > hard-drive/nic/display adapter needed. I might have something like that (a Pentium 100, actually), but if I give it to you, what will I do with all the CD-ROM drives and cards you'll give me? :) -- Nadav Har'El| Sunday, Feb 23 2003, 22 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A fine is a tax for doing wrong. A tax is http://nadav.harel.org.il |a fine for doing well. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: HAM radio (was: pcmcia cardbus lan)
On Tue, Feb 25, 2003, Arik Baratz wrote about "RE: HAM radio (was: pcmcia cardbus lan)": > > Try AES (www.aesham.com) or Universal Radio. AES has better prices, > > Universal has better service. > This is completely off-topic (as this thread was from the start...), but I am curious: Many years ago, ham radio was a very interesting way to "meet" people from other countries, chat with them, and get to know the world without living your home and without having your phone bill sky-rocket. These days, what does ham radio give you that the Internet doesn't, much more conveniently and without needing any licenses or tests? -- Nadav Har'El| Tuesday, Feb 25 2003, 23 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Early bird gets the worm, but the second http://nadav.harel.org.il |mouse gets the cheese. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Software design document
On Tue, Feb 25, 2003, Muli Ben-Yehuda wrote about "Re: Software design document": > On Tue, Feb 25, 2003 at 06:30:52PM +0200, kfir lavi wrote: > > > what is the best for complicated mathematical notations ? > > Tex, LaTeX, LyX, no contest. There's also "Texmacs", what supposedly is cross between TeX and Emacs, but in reality bears no relation to either (it's a WYSIWYG word processor, that does not actually use TeX - it only tries to look as good). Texmacs is a GNU project, or so they claim - check out http://www.texmacs.org/. Note that I never used texmacs, so I don't know how well it works in real life usage. My personal preference is using LaTeX straight, but I admit that it's not an appealing concept for someone who's already a Microsoft-Office junkie. -- Nadav Har'El| Wednesday, Feb 26 2003, 24 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |When you handle yourself, use your head; http://nadav.harel.org.il |when you handle others, use your heart. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Software design document
On Tue, Feb 25, 2003, Muli Ben-Yehuda wrote about "Re: Software design document": > > is the conversion to pdf is easy, with no faults? > > Yes. Let me know if you want to see my LaTeX makefile, which is based > on Oleg's. Also, make sure to follow the advice at > http://www.advogato.org/person/ladypine/diary.html?start=37 for > producing a valid pdf. Or, you might say "Hey, Adobe, you got it right the first time around!" and just use Postscript. Say no to helping Windows-users stay with Windows! [1] [1] http://www.fefe.de/nowindows/ -- Nadav Har'El| Wednesday, Feb 26 2003, 24 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I have a great signature, but it won't http://nadav.harel.org.il |fit at the end of this message -- Fermat = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Software design document
On Wed, Feb 26, 2003, Eli Segal wrote about "Re: Software design document": > Just one thought ... > Why would anyone want users to leave windows .. > > I say, if someone is happy with his windows .. let him be > linux is alternative and that's the good part The bigger issue that (I believe) is raised in that link I gave is that why should we, people who like Linux and develop for it, want to waste our time to help Windows users. For example consider someone who loves LaTeX but writes in Word format to please other (Windows-using) people. Or someone for whom postscript is most convenient but generates PDF anyway, because he believes Windows-users prefer it. Or some free software developer that spends (say) 10% of his development time to make sure his program works under Windows, rather than spending that 10% on something really important to him (assuming this developer himself does not user Windows and is not planning to use it anytime soon). This is obviously a very fanatic view of things, and I don't completely agree with it. But he does make a valid point. -- Nadav Har'El| Wednesday, Feb 26 2003, 24 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |God created the world out of nothing, but http://nadav.harel.org.il |the nothingness still shows through. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)
On Thu, Feb 27, 2003, Hetz Ben Hamo wrote about "Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)": > > I use "Mozilla 1.3a: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; > > rv:1.3a) Gecko/20021212" > > > > I looked at two pages because they show different behavior: > > > > 1. http://www.mechon-mamre.org/c/ct/c0101.htm > >with Hebrew only > > Looks perfect on Konqueror. I hope the win32 version of the KHTML version will Perfect?? For me, it couldn't look worse. I see no Hebrew letters or niqqud, just squares... I'm using Konqueror on Redhat 8.0 (kdebase-2.2.2-1). Maybe a later release of Konqueror fixed this issue? > > MS Internet Explorer 5 displays both pages correctly (with the same > > default font). Not that I give MS the credit for this, it might be > > the only browser the page-authors tried ;-) > > Probably. Mozilla in Linux shows it very bad. In Linux Mozilla, the niqqud appears to the side, rather than below, the letters. I believe this is a Mozilla problem, not a font problem, as I don't believe current X fonts even have the capability to define the correct positioning of Niqud (please correct me if I'm wrong). The niqqud-on-the-side happens in my Mozilla (1.0.1) with either pages that were linked (with or without the English translation). I get exactly the same problems in Arabic, so it's not a Hebrew-specific issue. -- Nadav Har'El|Thursday, Feb 27 2003, 25 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I don't eat snails. I prefer fast food. http://nadav.harel.org.il | = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)
On Thu, Feb 27, 2003, Meir Kriheli wrote about "Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)": > > Perfect?? For me, it couldn't look worse. I see no Hebrew letters or > > niqqud, just squares... I'm using Konqueror on Redhat 8.0 > > (kdebase-2.2.2-1). Maybe a later release of Konqueror fixed this issue? > > kdebase-2.2.2 in RH8 ? Are you sure ? Oops, sorry, I accidentally tested this on a Redhat 7.2 (!), on which I had kdebase-2.2.2. I will check it on Redhat 8.0 later tonight. If Konqueror can show Niqqud, it's great news :) -- Nadav Har'El|Thursday, Feb 27 2003, 25 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |What's tiny, yellow and very dangerous? A http://nadav.harel.org.il |canary with the super-user password. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)
On Thu, Feb 27, 2003, Nadav Har'El wrote about "Re: Niqud in Mozilla (was Re: need tool for high quality typeset...)": > Oops, sorry, I accidentally tested this on a Redhat 7.2 (!), on which I > had kdebase-2.2.2. I will check it on Redhat 8.0 later tonight. > If Konqueror can show Niqqud, it's great news :) I just looked at it on Konqueror from Redhat 8.0 (kdebase-3.0.3-14), and indeed the Hebrew looks all right, but the niqqud is completely absent... :( http://www.mechon-mamre.org/c/ct/c0101.htm -- Nadav Har'El|Thursday, Feb 27 2003, 25 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Support bacteria - they're the only http://nadav.harel.org.il |culture some people have! = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: TIP: using CVS keyword in LaTeX documents
On Sun, Mar 02, 2003, Muli Ben-Yehuda wrote about "TIP: using CVS keyword in LaTeX documents": > I was looking for a way to do this for quite sometime. The problem is > that LaTeX interprets the '$' in $Id$ as "enter math mode" and thus > doesn't render the CVS Id string appropriately. A nice hack is to do: > > $ $Id$ $ > > Works like a charm! Another trick is (and I didn't check...) \verb$Id$ or if you want to see the "$"s in the output, \verb+$id$+ -- Nadav Har'El| Sunday, Mar 2 2003, 28 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |"Outlook not so good." Wow! That magic 8- http://nadav.harel.org.il |ball knows everything! So, what about IE? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: TIP: using CVS keyword in LaTeX documents
On Sun, Mar 02, 2003, Muli Ben-Yehuda wrote about "Re: TIP: using CVS keyword in LaTeX
documents":
> > \verb$Id$
> True. However, \verb apparently can't be used as a command
> argument. See http://www.ecn.purdue.edu/users/notz/latex-cvs.html.
>
> So, any other cool latex tricks? Or do I have to open latex-il first?
Doesn't \protect fix that? E.g., try
\chapter{\protect\verb+$Id$+}
(again, I didn't try it).
\protect is necessary in some cases of *moving* arguments (arguments that
get copied to other places, like table of contents, footnotes, etc.), and
I consider its existance a LaTeX bug, not a feature.
--
Nadav Har'El| Sunday, Mar 2 2003, 28 Adar I 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |A man with a watch knows what time it is.
http://nadav.harel.org.il |A man with two watches is never sure.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: TIP: using CVS keyword in LaTeX documents
On Sun, Mar 02, 2003, Oleg Goldshmidt wrote about "Re: TIP: using CVS keyword in LaTeX documents": > I also suppose you don't need spaces: $$Id$$ will work... Actually, "$$" is TeX for "start an equation on a seperate line", so it won't work as you expect. -- Nadav Har'El| Sunday, Mar 2 2003, 28 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The trouble with political jokes is they http://nadav.harel.org.il |get elected. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Equipment
On Sun, Mar 02, 2003, Voguemaster wrote about "Re: Equipment": > Damn, I could use an ISA soundblaster... hehe > Couldn't find one anywhere! So now all you need are three camels! Then you could make the following series of crazy trades, paraphrasing on various emails people sent here recently: three camels -> seven sheep and one duck -> a Commodore 64s and two Apple II -> one microvax with a Unix source license -> two 386sx motherboards and 1K of original core memory -> ten pounds of original Hollerith punch cards -> an ISA soundblaster. Enjoy the soundblaster (and don't forget to clean up after the livestock!) -- Nadav Har'El| Sunday, Mar 2 2003, 28 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The trouble with being punctual is that http://nadav.harel.org.il |nobody is there to appreciate it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [Which list?] Unicode, bidi, terminals and tables
On Mon, Mar 03, 2003, Beni Cherniavsky wrote about "[Which list?] Unicode, bidi,
terminals and tables":
> I'm looking for the Right Model of terminals that would support bidi
> well, with minimal hassles to the application. In particular, it's
> imperative that `cat foo.txt' does the right thing. I don't care much
Are you sure you know what the "right thing" for cat foo.txt to do is?
It's more complicated than it sounds, especially when you're talking about
mixed-language text (e.g., Hebrew and English). Do you want the "main
direction" to be always LTR? be figured out per-line (as, for example,
QT seems to do)? Per paragraph?
My "bidiv" program attempts to be such a "cat"-like program which works
on normal non-bidi terminals (i.e., it does the bidi itself). Its default
heuristic is to decide the direction per-paragraph, but I'm very skeptical
that a terminal emulator can use this kind of heuristics without annoying
a lot of people.
> for the existing vt100 model; I'd actually like to take the
> opportunity and throw most of its complexity out of the window but
> that's not a requirement, the bidi is.
Most of the vt100 "complexity" is used by cursor-addressed programs, like
for example vi and mutt. Some redundant looking escape sequences ("e.g.,
delete from the cursor to the end of line") are used to lower the number of
characters that need to be output to modify a screen (libraries like "curses"
specialize in using this efficiently), and was very important in the
days of slow modems (remember 110 baud? :)) - and to some degree it's still
important. I don't know why you'd want to get rid of these features,
especially as they already exist (e.g., in "xterm").
> The model I'm thinking of is based not on a visual-ordered matrix of
> characters but rather on a logical stream of text. The terminal
Yes, what you describe is a teletype (see
http://www.columbia.edu/acis/history/teletype.html) - printers used
before the advent of CRT terminals, which did not have cursor addressing.
Do you think anybody wants to return to this?
Feel like going back to "ed" for editing? :) I certainly don't.
That being said, if you believe that cursor-addressing applications are
"passe", and people only use graphical applications (openoffice, emacs,
mozilla, etc.) nowadays, and the xterm window is only used for commands
like "cd", "ls", "cat", your idea might have merit. But with you're
permission, I'll disagree.
> * They assume fixed-width display. This requires fixing to use
> wcwidth() and still is fragile. Perhaps it should be replaced by
> some convention of signalling the terminal "align this under this"
> that does not require the application to count anything at all.
Remember tab stops? This is what they were for :)
> every cell to emit the correct marks. It might be acceptable to
> expect `sdiff' to output some markers but it's unreasonable to
> expect it to check the first strong char in every line! It'd be
Maybe the conclusion is that an application which wants to diff
multi-lingual multi-directional text needs to be a little smarter then
sdiff, and it's not an issue of terminal emulator at all?
And what about diffing texts which contains left-right English, right-left
Hebrew, and top-down Japanese? It seems to me that some problems are not going
to be solved in the terminal-emulator level.
> So there is a need for extending "plain-text" unicode with some extra
> semantics that will allow me to express a table (at least sa far as
> bidi should work). I found some old discussions on the unicode
Plain text should not have tables.
Higher-level text files, like HTML, TeX or Troff sources, XML OpenOffice
files, etc., do contain tables and have their own mechanisms for specifying
the language in each column and so on.
> mailing list that lead nowhere. I don't aim at converting the whole
> world t ouse these extensions; I just want something that can be used
> at the command line in interactions between different unix utils (just
> like there is no end-of-line concensus but unix works happily with
> \n).
If you produce something that has real-world usefulness, I'll use it :)
Note that what you described does not need to be a terminal-emulator at
all! It can, and perhaps even should, be simply a pseudo-terminal layer
which processes the text and passes it to a normal, bidi-agnostic,
terminal emulator. I believe I once saw the Arabeyes project have exactly
such a system (I don't remember its name).
> I want to be able to `cat table.txt'; having to use visual apps for
> that is a no-no.
But why is "bidiv table.txt" a no-no?
If you
Re: [Which list?] Unicode, bidi, terminals and tables
On Mon, Mar 03, 2003, Nadav Har'El wrote about "Re: [Which list?] Unicode, bidi, terminals and tables": > Note that what you described does not need to be a terminal-emulator at > all! It can, and perhaps even should, be simply a pseudo-terminal layer > which processes the text and passes it to a normal, bidi-agnostic, > terminal emulator. I believe I once saw the Arabeyes project have exactly > such a system (I don't remember its name). I think what I had in mind was Acon, or Akka (the latter is a rewrite and supposedly improvement of the former). Check out arabeyes.org for more information. -- Nadav Har'El| Monday, Mar 3 2003, 29 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |"I'll doublecross that bridge when I come http://nadav.harel.org.il |to it" (a politician about the future) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [Which list?] Unicode, bidi, terminals and tables
On Mon, Mar 03, 2003, Beni Cherniavsky wrote about "Re: [Which list?] Unicode, bidi, terminals and tables": > > I think what I had in mind was Acon, or Akka (the latter is a rewrite and > > supposedly improvement of the former). Check out arabeyes.org for more > > information. > > > Great, didn't know such a thing already exists! That should allow > bidi in the console. Will grab it as soon as I recover my HD :-). Another thing you might want to look at is luit (see http://www.pps.jussieu.fr/~jch/software/luit/, and it also comes by default in Redhat 8). Luit does not currently implement bidi (the guy who wrote this discusses the reason in the aforementioned page) - but you might be interested to look at its code, if you plan to write such a filter yourself. -- Nadav Har'El| Monday, Mar 3 2003, 29 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |In God we Trust -- all others must submit http://nadav.harel.org.il |an X.509 certificate. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [Which list?] Unicode, bidi, terminals and tables
On Mon, Mar 03, 2003, Beni Cherniavsky wrote about "Re: [Which list?] Unicode, bidi, terminals and tables": > - The raw/cooked modes are not very elegant; cooked mode actually > belongs in user-space. 9term had an elegant idea of user-trigered Well said, though the invention of "ptys" in modern Unix systems (and Linux, of course) solved this many years ago. Applications like screen, ile (that adds history-support to any text application, say, bc), akka, use this to sit between the graphical terminal emulator and the final application. > * Termcap/info - I'm not sure this is needed; if a single set of > escapes defined once (or with revisions) will do, that's prefered > (but I do see the future compatibility value of such a thing). This is probably what every person who invented a terminal thought :) Termcap/terminfo exists because it is not true, and people still use various terminal emulators for different reasons, each implementing different sets of escape sequences. > Since I want to make creating terminal applications easier, I want to > minimize the entry cost of working with bidi properly. So expecting > full-screen programs to disable implicit bidi and do it all by > themeselves is out of the question. > > The best current model is that the application edits a rectangular > matrix of chars (thinking that's a vt100) that is rendered through > line-by-line implicit bidi. This breaks as soon as the application > puts texts side-by-side that are not logical continuations. I don't > think dialogs will survive it if they contain mixed hebrew/english > text. The worst scenario is a pop-up window covering some of the > mixid text below it. If at least the content of the pop-up window > will be in place, that would already be a miracle -:). > > That's why I'm unsatisfied with this model. The terminal has no idea > about the nesting of 2d areas of the application, so it can't help it > with the bidi. One way to work around this problem is to add the bidi support into the display library, i.e., Curses. Most cursor-addressing applications (at least, the ones that originally came with Unix) are written using curses, which takes care of supporting "windows" on the actual terminal window, and in outputting the changes to the display in the most efficient (and correct) way possible. Curses might be, conceivably, be modified to support bidi. It already knows of seperate unrelated windows, and it can be perhaps be modified to support unicode and bidi better (I have to admit that I have no idea how well curses currently supports them). > > The appropriate mailing list is, in my opinion ivrix-discuss. > > > Where are the 2003 archives ?_? Is it alive? It's alive. In a few minutes, I'll finally change the link on ivrix.org.il, but in the meantime, here are the links to the archives: http://news.gmane.org/thread.php?group=gmane.linux.region.israel.ivrix.discuss (nice interface to the last few months) or http://www.math.technion.ac.il/lists/ivrix-discuss/ (the full archives, but in ugly monthly format...) -- Nadav Har'El| Monday, Mar 3 2003, 30 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |When you lose, don't lose the lesson. http://nadav.harel.org.il | = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: egged.co.il works
On Tue, Mar 04, 2003, Tzafrir Cohen wrote about "Re: egged.co.il works": > So another broken site is supported, instead of it being built properly. Right... And worse, this gives Microsoft a whole new baseline on which they can create new divergent technology, which gets further and further from W3C standards, while Konqueror continues to chase its tail... This is why Netscape freed the Mozilla sources - to stop Microsoft from being able to do that. And this is why Mozilla has very strong opinions on not emulated broken IE behaviour. -- Nadav Har'El| Tuesday, Mar 4 2003, 30 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Life is what happens to you while you're http://nadav.harel.org.il |busy making other plans. - John Lennon = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: egged.co.il works
On Tue, Mar 04, 2003, [EMAIL PROTECTED] wrote about "Re: egged.co.il works": > Organize? Maybe create a site or a subsite which contains links of all the > uncooperative sites, and give them bad rap in the press? Create a nice PDF in > Hebrew about standardization of a site, and send it to management levels in > those companies? I'm at a loss. For me it's worse than most of you - my other OS > is a MacOS... Maybe it's time we (any lawyer in the house?) try to find out whether any of these sites actually break any laws - e.g., by virtue of discriminating against consumers based on irrelevant criteria, descriminating against blind consumers, and so on. If it's not illegal, it still stinks, and maybe we can get some pro-consumer organization or newspaper, or something, on this issue. Companies should be made to understand that making their site look "less nice" in Mozilla is perfectly acceptable, but making them completely inaccessible because of stupidity - while all the W3C standards are all with strong emphasis on backward compatibility and accessibility - is simply inexcusable. -- Nadav Har'El| Tuesday, Mar 4 2003, 30 Adar I 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Spelling mistakes left in for people who http://nadav.harel.org.il |feel the need to correct others. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: egged.co.il works
On Tue, Mar 04, 2003, Shachar Shemesh wrote about "Re: egged.co.il works": > me focuse my efforts. Please give a quick vote of the site whose lack of > support for konq/moz is the most troubling, and let me know. I think > Bank Leumi is the candidate for this. Anyone else? Yes. Bank Leumi is the worst as far as I'm concerned. -- Nadav Har'El| Tuesday, Mar 4 2003, 1 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Today is the tomorrow you worried about http://nadav.harel.org.il |yesterday, and now you know why. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Browsers, compatility, and price
On Tue, Mar 04, 2003, Hetz Ben Hamo wrote about "Browsers, compatility, and price": > As my previous boss in Tehila told me: > "Hetz, there are 2 ways to choose from: > > 1. A repsectable company who will build the site with PHP, MySQL and will > fully check that their web site will work on various browsers - those > companies ask for X*2 price > 2. A small company who will build the site on IIS, with ASP and will only > check it against latest version of Explorer - they will ask for X price.." > > Guess who wins the bid? Who said these are the only two option??? Maybe your boss was simply ignorant? What happens if you take the cheaper company and tell them "don't do me any favors - don't write javascript, don't write any flash animations. I want every link in the page to be an HTML ". What do you think will happen? They'll increase their price? I seriously doubt that. Their work will actually become easier! > 1. Konqueror, Opera - these browser do whatever they can to support what > Opera's CEO calls "streets HTML" - which means either non standard HTML or > buggy behaviors - both browsers support those kind of HTML in addition to the > W3C standards - that way you can browse many problematic sites. And what about support for Microsoft-invented Javascript? And ActiveX? And if Konqueror and Opera support this, what will prevent Microsoft from coming up with more crazy changes later? Remember, it's exactly those incompatibilities that cause Microsoft users to upgrade their OS and applications. Try to stick with "MS Word 95", and see if the Windows-junkies don't laugh at you just as hard as they laugh at Linux users. > Financially speaking - writing a web site and maintaining it work with the > latest variant browsers is a PITA. You got tons of quirks to do in order to > make the site look good on MSIE and Mozilla, and that part alone sometimes To look good, yes, it's a PITA. To make it *work* however, is not. In fact, it takes a lot of effort to make it NOT WORK. I still can't believe some people (like the boss you described) don't get it... Note that sites like Egged's or Bank Leumi are meant to be used, not admired (unlike, for example, a site advertising some movie which you are supposed to go out and see in the cinema), so if the people in charge there asked for "cute" features like sophisticated Javascripts - they completely didn't understand the job they were asked to do. -- Nadav Har'El| Tuesday, Mar 4 2003, 1 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Make it idiot proof and someone will make http://nadav.harel.org.il |a better idiot. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [Fwd: BSA Accuses OpenOffice ftp sites of piracy]
On Tue, Mar 04, 2003, Eli Billauer wrote about "Re: [Fwd: BSA Accuses OpenOffice ftp sites of piracy]": > But the problem is not that some people want a WYSIWYG word processor, > or spreadsheets. Micro$oft doesn't have any rights on the WYSIWYG > concept. But if you call an application KWord and make it look like > MS-Word, you've made a statement about its origins. Excuse me? M-w.com lists the phrase "word processor" as dating back to 1970, which is a good 15 years before the first version of MS-Word I know of. Microsoft has no copyrights on the generic English word "word", and while calling a word processor, of all names, "Kword" may be strange - it is certainly not illegal. And about "looking like MS-Word", do you also say that Mazda cars (say) stole something from Ford, because Ford had the concept of a car, with four wheels and an engine, before Mazda? Since when is honestly producing competing products to successful products illegal, or even immoral? Next thing we know, MS will be chasing offices not using MS-Office, because, obviously "You can't call yourself an office if you don't run MS-Office!" :) -- Nadav Har'El| Tuesday, Mar 4 2003, 1 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Sign in zoo: Do not feed the animals. If http://nadav.harel.org.il |you have food give it to the guard on duty = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: egged.co.il works
On Wed, Mar 05, 2003, Guy Baruch wrote about "Re: egged.co.il works": > > > Nadav Har'El wrote: > > >Yes. Bank Leumi is the worst as far as I'm concerned. > > > > > but bank leumi has a workaround as the old site, which was posted here > some months back. No, the workaround stopped working a week after you posted it. It now uses an invalid certificate (!) to secure the connection, which makes the site unusable. > let's not open that old discussion again, but if it has a WA, it's not a > show stopper ... A non-secure workaround is as good as no workaround, in the case of banking... -- Nadav Har'El|Wednesday, Mar 5 2003, 1 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The meek shall inherit the Earth, for http://nadav.harel.org.il |they are too timid to refuse it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Submitting Debs
On Wed, Mar 05, 2003, Eli Segal wrote about "Submitting Debs": > Hey .. > > Does anyone here know, where can i find information > on packing .debs and submiting them ??? > > I want to pack various of software in hebrew and submit them > so we could just install rather then compile'em Baruch Even is already doing very good work in this direction, and has quite a number of Hebrew packages in Debian unstable, including hspell, bidiv and culmus (to quote just a few I remember). So you might want to contact him directly to reduce duplication of work. -- Nadav Har'El|Wednesday, Mar 5 2003, 1 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Lumber Cartel member #2224. http://nadav.harel.org.il |http://come.to/the.lumber.cartel = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: A problem with the CD media or with the CDROM drive?
On Sat, Mar 08, 2003, Shaul Karl wrote about "A problem with the CD media or with the CDROM drive?": > Would you say the following points to a media failure, that is a > faulty production of this specific disc? > > $ ls -l /cdrom/KNOPPIX/KNOPPIX > -r--r--r--1 root root 717842384 Jan 18 19:36 > /cdrom/KNOPPIX/KNOPPIX This is a very large image... Could it be that this blank was only meant for smaller image, say 650 MB? If "MB" means 1000K (it many times does in the disk industry...), then this image is even larger than 700MB.. -- Nadav Har'El| Saturday, Mar 8 2003, 4 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A man is incomplete until he is married. http://nadav.harel.org.il |After that, he is finished. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Submitting Debs
On Sun, Mar 09, 2003, Shaul Karl wrote about "Re: Submitting Debs": > On Sun, Mar 09, 2003 at 10:36:44AM +0200, Shachar Shemesh wrote: > > > > > Email the relevant maintainers. > It is my opinion that submitting bug reports, for example with the > reportbug utility, is better. 2 reasons for this are: But before reporting bugs, please verify that they indeed exist. For example, > > apt-cache show licq > > ... > > Maintainer: Zed Pobre <[EMAIL PROTECTED]> > > ... For me (on Redhat 8.0), licq works very well with Hebrew, thanks to the support in QT. It also worked well with Hebrew in one or two earlier versions, for at least an year now.. So before you go complaining to the licq maintainer, please check that your Debian does not include 2-year-old copies of QT or Licq. And use the QT or KDE "plugins" of licq (I doubt that Hebrew support will work just as well in the other plugins). And make sure you have Hebrew keymap working (without it, you won't be able to type any Hebrew, not in licq, not in mozilla, and not in most other applications, but you will be able to see Hebrew.) -- Nadav Har'El| Sunday, Mar 9 2003, 5 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |There are 2 ways to do it - my way and http://nadav.harel.org.il |the right way = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Insta Party!
On Tue, Mar 11, 2003, Shachar Shemesh wrote about "Re: Insta Party!": > Oron Peled wrote: > > >It is extremly important, especially with respect to security > >updates. We don't want to create extra 100 vulnerable sendmails > >out in the wild now, do we? > > > No reason to have a listening sendmail on the installed machines. We > install machines that have 0 (zero, nada, zilch, gurnisht, mafish) > listening services, and the chances of exploitation drop tremendiously. Indeed, Redhat starting from version 7.1 (if I remember correctly), have a quite "safe" installation by default, including a sendmail only listening on localhost and other crap (telnet, ftp, etc.) disabled by default. But users, especially newbies, might be tempted to turn these on (wow, I can have a ftp server, web server and mail server on my system connected through a modem? cool!), so they better not have holes... > AFAIR licq had vulnerabilities as well, so this does not eliminate the > risks totally, but it greatly reduces them. Yes, there were a number of holes in client programs (even in stuff like the "file" command...), libraries (openssl), and so on. By the way, I wonder if it's possible to instead of creating an update CD, simply replace the updated RPMs on the original Redhat installation CDs... -- Nadav Har'El| Tuesday, Mar 11 2003, 7 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |"Mommy! The garbage man is here!" "Well, http://nadav.harel.org.il |tell him we don't want any!"- Groucho Marx = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Insta Party!
On Tue, Mar 11, 2003, Ira Abramov wrote about "Re: Insta Party!": > Quoting Alon Altman, from the post of Tue, 11 Mar: > > The CDs have already been sent to be copied, and the updates take > > almost a full CD by themselves. > > and this is why, girls and boys, I would never touch a *.0 Red Hat > edition with a 30 meter pole. Ira, you're completely misdirected on this issue. Almost all of these "errata" are because of actual vulnerabilities discovered in software included in Redhat, not some "*.0" illnesses. Just to name the most recent serious incident, sendmail was updated after a remote vulnerability was discovered. Updates exists for Redhat 8.0, but also for Redhat 7.3, 6.2, and (yes!) Debian. Continuing to use a 1 year old distribution does *not* save you from the need to continuously upgrade when security flaws are found. Besides, as I explained once and again, newbies *must* be given only the *latest* release available, because such a release will always support newer hardware better, and support Hebrew better. If you give someone Redhat 7.3, the next thing you know this list will be flooded with questions like "I'm trying to upgrade to KDE ... for the improved Hebrew but I'm having troubles", "X doesn't support my video card", "Mozilla has bugs in Hebrew", and so on... > RHL8 has been out less than half a year, and has caused: > > functionality breaks due to beta builds > functionality headaches due to Unicode incomplete support in apps > functionality slowdown due to Unicode bugs (grep for instance) I don't know anybody that this ever bothered... On the other hand many people are quite pleased by the better support for Unicode... Maybe you call this support "premature" because you yourself chose not to use it, perhaps because you use English almost exclusively on your system? > and now you tell me the updates themselves take up about 600 meg? > youch! The updates are simply whole packages that were replaced. It includes the entire KDE, mozilla, apache, kernels - so no wonder it takes about 500 MB... > better install Debian sid, IMSBO. got newer software but miracleously > better stability, and none of that premature Unicode voodoo. As a Debian user in the last few months, I'd like to disagree. Debian was not more stable than Redhat (e.g., we had some glibc problems that I never saw in Redhat), was slower in sending out updates (e.g., to close the sendmail hole). It certainly wasn't bad, but I wouldn't say it wins over Redhat with one hand behind its back. -- Nadav Har'El| Tuesday, Mar 11 2003, 7 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The two rules for success are: 1. Never http://nadav.harel.org.il |tell them everything you know. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Announce: Hspell 0.4
On Tue, Mar 11, 2003, Muli Ben-Yehuda wrote about "Re: Announce: Hspell 0.4": > On Mon, Mar 10, 2003 at 08:58:19PM +0200, Dan Kenigsberg wrote: > > On Mon, Mar 10, 2003 at 12:15:04PM +0200, Matan Ninio wrote: > > > I'm very happy to here this. I would be even happier if I would > > > understand what I need to do to get this to work with my bidi-emacs. > > > Have you looked into the possibility of interfacing with emacs ispell > > > like? being the center of the computational creation, the One True > > > Editor is an excellent platform to attach to, if you want to make this > > > speller useful. > > > Matan > > > > The glove is thrown down. Who will dare to pick it up? > > I'd be willing to look into it, time permitting. Matan, a pointer to > the bidi-emacs sources? or are you referring to plain MULE? I doubt the spell-checking elisp package has anything to do with whether that Emacs supports bidi or not, so I suggest you start with any-old-Emacs- that you already have installed. I tried myself to make hspell work with ordinary Emacs (XEmacs is my favorite variant, actually), because I promised Richard Stallman I'd do that, but I couldn't get it to work; "ispell.il" runs "ispell -a" to do the actual spell- checking, and it can be easily pursuaded to run "hspell -a" instead. However, there was something in hspell's replies that it didn't like, which is quite understandable, because hspell -a's output doesn't exactly conform with what it should have been (it's the minimum needed to get LyX to work with it). So supporting hspell in Emacs would probably entail finding what bothers Emacs in hspell's output, and fixing it in hspell (the latter is probably simple, as hspell.pl is a relatively simple perl script). Muli, if you try doing this and run into any problems, feel free to ask me or Dan for help. -- Nadav Har'El| Tuesday, Mar 11 2003, 7 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Strike not only while the iron is hot, http://nadav.harel.org.il |make the iron hot by striking it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Insta Party!
On Tue, Mar 11, 2003, Ira Abramov wrote about "Re: Insta Party!": > Quoting Nadav Har'El, from the post of Tue, 11 Mar: > > Besides, as I explained once and again, newbies *must* be given only the > > *latest* release available, because such a release will always support > > newer hardware better, and support Hebrew better. If you give someone > > well, across linux-il and another LUG list I'm on, very User who tried > it complained. I haev yet to hear ONE person, other than Hetz, say a > good word about RHL8. Ave Redhat 8.0, linuxuri te salutant! Is that what you wanted to hear? I use Redhat 8.0 and I'm very pleased with it. It is not flawless, and has a few bugs, but it is still better than any Redhat release that came before it. The updated versions of all the packages (yes, even gcc), the inclusion of OpenOffice, improved Konqueror, and many other new features in this release more than made up for any new bugs that may have crept in. > why not use Mandrake if Hebrew is so important? How is Mandrake better than Redhat in that respect? I am guessing that the latest Mandrake is just as good as the latest Redhat, but not much (or even any) better. (but this is just a guess - I never used Mandrake myself). > if grep slowed down by 3(!!) orders of magnitude, you would notice, I > promise you. It was one example I read on a list, I haven't installed it I promise you, I never noticed such a slowdown. And you don't have to work in the UTF8 locale if you don't want to, it's just an option... I continue to use ASCII :) > (or whatever it was the preceded it) by including all the updates and > crypto (before it was built-in) into the main install list. One thing I > noticed that RPM can't do to this day is to survive (cd RPMS ; rpm -Fvh > *) if there are too many files in the directory. it may die and corrupt > the database. just one more reason I like apt-get better. There's is indeed a bug in Redhat 8.0's "rpm" command which sometimes hangs for no reason. It doesn't consistently happen, and apparently is not specific to large directories, and happens randomly (repeating the same command works). Luckily, it happens very rarely for me. -- Nadav Har'El| Tuesday, Mar 11 2003, 7 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I want to live forever or die in the http://nadav.harel.org.il |attempt. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: small problem with mount
On Tue, Mar 11, 2003, Alon Barzilai wrote about "small problem with mount": > #mount /dev/hdb1 /mnt/tmp > I get the error message: > fs type ntfs not supported by kernel > however, > #mount -t ext3 /dev/hdb1 /mnt/tmp > works. > > is there anything else I have to make so the disk will not remember it's > ntfs past ? You probably forgot to use "fdisk" to change the partition type byte on that partition. Do "fdisk /dev/hdb", and there do 'l' and then 't', changing the partition type of that partition to 83 (Linux). Also, it is recommended to add a line for /dev/hdb1 in /etc/fstab, with something like /dev/hda1 /mnt/tmp ext3defaults1 5 -- Nadav Har'El| Tuesday, Mar 11 2003, 7 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Boat: A hole in the water surrounded by http://nadav.harel.org.il |wood into which one pours money. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: egged.co.il works
On Wed, Mar 12, 2003, Oleg Goldshmidt wrote about "Re: egged.co.il works":
>...
> Ultimately, the people needed to be convinced are decision-makers, not
> techies. Decision-makers can only be convinced by money arguments. If
>...
> Coming back to Shachar's original request for ammunition, please give
> him economic arguments in favor of standard compliance. No other
> arguments are relevant in the context, IMHO.
This is a good point, but not entirely true. Issues of discrimination
("blind people can't use your site", "10% of the Israelis can't use your
site", "Bank Hapoalim doesn't descriminate thus!") and future certainty
("the moment IE 7 comes out, your site *may* stop working!") may also mean
something to the honest, forward-looking, manager.
Of course, to a greedy manager with plans to keep his office for only the
next year and caring just about short-term profits, such arguments will not
be relevant.
--
Nadav Har'El| Wednesday, Mar 12 2003, 8 Adar II 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |The trouble with political jokes is they
http://nadav.harel.org.il |get elected.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: pipe question
On Fri, Mar 14, 2003, Tzafrir Cohen wrote about "Re: pipe question": > tee to a file >.. > I also recall a "multi-cat" or "multi-pipe" file, but can't seem to find > it anywhere. > > Also: can this be done using a more complicated redirection in a > standard shell? With Zsh (my favorite shell, and works on every conceivable system including Linux and Windows), you could do, with the "multios" feature: setopt multios ls >file | grep a # saves ls output on file, and also greps "a" ls >file1 >file2# saves ls output to both file1 and file2 This might look like a silly way to save an extra "tee" run, but an external "tee" program sometimes causes annoying side-effects that ZSH's multios solves (not to mention it being more convenient). For example, if you do ls | tee file1 >file2 The exit code of that pipeline is that of tee (i.e., normally a success) and you don't know whether or not ls succeeded. With multios, you get to know the ls exit code: ls >file1 >file2 By the way, the "multios" option is not enabled by default, perhaps because it's a bit dangerous - if you do echo hi >* You'll overwrite every file in the current directory with "hi" :( -- Nadav Har'El| Friday, Mar 14 2003, 10 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A Nobel Peace Prize? I would KILL for one http://nadav.harel.org.il |of those. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pipe question
On Fri, Mar 14, 2003, shlomo solomon wrote about "Re: pipe question": > So what's my problem? It seems that both the above solutions are writing to a > buffer and the actual screen output is not immediate, but in spurts. Let me > make that clearer. The original script creates a few hundred lines of output > that would normaly (if not for the pipe and GREP) scroll **gently** off the > screen during the few minutes it runs. But with both the above solutions, I > get nothing on screen at first. After a little while, about 100 lines scroll > at high speed and a while later, another 100, etc. This is certainly better This is happening because of a convention, an implementation detail, of the standard C library. By convention, the "stdout" file pointer is line-buffered, *unless* stdout is redirected not to a terminal (a pipe or a file) in which case this file pointer becomes fully buffered, meaning that output is not flushed by default until full buffers (e.g., 4096 characters) are accumulated. The C library checks whether stdout is pointing to a terminal using isatty(). This is why grep appears to output line by line while grep ... > file writes to the file in chunks. If you have access to the source-code of the program, you can add a setlinebuf() call to make stdout line buffer, or add fflush() calls. But since I assume you don't want to do that, you can use the "unbuffer" utility on an existing program, e.g., unbuffer grep > file and it causes grep to think that its output is still a terminal, so it remains line-buffered. On my system (Redhat 8.0), "unbuffer" is implemented as a trivial (3-line) "expect" stript; If you want an interesting exercise try writing a C version using a different approach: set LD_PRELOAD to a shared library containing a fake implementation of isatty(), one that always returns 1 (or returns 1 for fd 1 and calls the original function for the others). -- Nadav Har'El| Friday, Mar 14 2003, 10 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The meek shall inherit the Earth, for http://nadav.harel.org.il |they are too timid to refuse it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pipe question
On Sat, Mar 15, 2003, Shaul Karl wrote about "Re: pipe question": > Am I right that you could have done it in C (as well as with other > languages) by duplicating the file descriptors, redirecting some of them > and so on? The reason for asking is that bash has some of these > capabilities: I don't think so... What you want is the same data to be written to do different files, or at least two different file descriptors. I don't know of any way of doing that besides actually calling write() twice to write the data to the two fds. This is what tee, or zsh's multios feature, does. Duplicating a file descriptor will get you two different file descriptors pointing to the same file - which is I think is the opposite of what the original poster wanted. -- Nadav Har'El| Saturday, Mar 15 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If you drink, don't park; Accidents cause http://nadav.harel.org.il |people. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Soon Announcements
On Sun, Mar 16, 2003, Shachar Shemesh wrote about "Re: Soon Announcements": > >Yetzi'a meavdut lekherut indeed. Who's organizing and where do we sign > >up? > http://www.pc.co.il/go-linux Say, does registration on that site actually *work* for anybody? I've been trying to register to that thing for almost two months - at first there was no "register" link, and recently there's a big registration form (forcing you to give them all sorts of possibly-irrelevant private details) but no "register" button at the end! My failure to register is especially silly given the fact that I'm supposed to present in that conference :) -- Nadav Har'El| Sunday, Mar 16 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Earth First! We can strip-mine the other http://nadav.harel.org.il |planets later... = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Soon Announcements
On Sun, Mar 16, 2003, Ariel Biener wrote about "Re: Soon Announcements": > On Sunday 16 March 2003 12:44, you wrote: > > Say, does registration on that site actually *work* for anybody? I've been > > trying to register to that thing for almost two months - at first there was > > no "register" link, and recently there's a big registration form (forcing > > you to give them all sorts of possibly-irrelevant private details) but no > > "register" button at the end! > > There is a register button at the end, but you can only "press" it using MSIE. Yeah, now that *REALLY* makes a lot of sense for a Linux conference :( I urge whoever in charge of this site to fix this problem quickly. I'm also starting to wonder what the fact that I'm the only person complaining about this issue in two months might mean about the number of Linux-savvy attendants to that conference :( > All the irrelevant details on the registration page are non compulsory, that > means you don't have to fill those out. That's not what I see there... I consider my ID totally irrelevant (yes, I understand why it's convenient for them, but it doesn't mean I agree to give it to them) and the address of my company, its fax number, and so on, are also something that should have been optional (some of us are unemployed, working for a company having nothing to do with our Linux hobby, or other bizarre but common situations). -- Nadav Har'El| Sunday, Mar 16 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |We are Microsoft. You will be http://nadav.harel.org.il |assimilated. Resistance is futile. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Soon Announcements
On Sun, Mar 16, 2003, Shachar Shemesh wrote about "Re: Soon Announcements": > >My failure to register is especially silly given the fact that I'm supposed > >to present in that conference :) > > > I wouldn't worry about that. I'm not sure we should be discussing this in the open like that, but I'll do it anyway and duck... I am worried about that, because I agreed to present in that conference under the impression that it was a serious one, organized by serious people (like you, Shachar) and attended by many people I'd like to meet. But when I see that people on this list don't know about this conference, that Mozilla users cannot even register (I've been trying for a couple of months! Most people would just leave the site and forget all about it.) I am worried. I already expressed both these worries a month ago, but they were not fully addressed. Obviously what I was worried about wasn't that they wouldn't let *me* in... That being said, I believe that this conference, hoping that this time they won't postpone it and that people manage to register, has the potential of being very interesting :) -- Nadav Har'El| Sunday, Mar 16 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Does replacing myself with a shell-script http://nadav.harel.org.il |make me impressive or insignificant? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Soon Announcements
On Sun, Mar 16, 2003, Oleg Goldshmidt wrote about "Re: Soon Announcements": > Well, they say outright that registration is only free for > "professionals" who work in the field. So it well may be that > "hobbies" don't count. Which is exactly why I thought of registering > as HaMakor emmber rather than just privately. When I said "hobby", what I had in mind wasn't pottery or macrame, but rather a deep interest that goes beyond your involvement in your current company. A large percentage of the relevant people for this conference are likely to be currently unemployed, switch jobs a month after the conference, currently employed for a company that does not have immediate involvement in Linux, or a company that doesn't like its employees to go to conferences, or whatever. So don't be surprised if people with IDs like 123456789 and working for "Vandalay Industries" or "Kramerica" [1] register to the conference ;) Using the name "Hamakor" will work just as well :) As they say, you can lead a horse to water but you can't make it drink - they can enforce entering these details but have no scalable way to actually verify any of them except your email and phone number. [1] These are Seinfeld references... :) -- Nadav Har'El| Sunday, Mar 16 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Why do programmers mix up Christmas and http://nadav.harel.org.il |Halloween? Because DEC 25 = OCT 31 = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Soon Announcements
On Sun, Mar 16, 2003, Gilad Ben-Yossef wrote about "Re: Soon Announcements": > If we have 500 people they will listen. If we are 25 people they wont. But what if those 25 people are half the lecturers they have planned for the event? :) -- Nadav Har'El| Sunday, Mar 16 2003, 12 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Unix is user friendly - it's just picky http://nadav.harel.org.il |about it's friends. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [Announcment] Hamakor is accepting new members & friends (and sells some CDs)
On Sun, Mar 16, 2003, Tal Peer wrote about "Re: [Announcment] Hamakor is accepting new members & friends (and sells some CDs)": > > a. Be 18 or older. > > Is this required by law? I would like to register as a member but seems > like i'll have to stay a friend for a few years :-) I am not a lawyer, etc., but there might be laws preventing minors from signing things like amuta membership forms. Of course, this might be wrong - after all we're talking about a country that has no problems taking 17 year old boys and asking them to sign papers in which they commit themselves to 6 years of army service... The Amuta's lawyer probably has a final answer to this question. -- Nadav Har'El| Sunday, Mar 16 2003, 13 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Always remember you're unique, just like http://nadav.harel.org.il |everyone else. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Somebody mickeyed with my mouse?!
Today I upgraded my PC at work. The hardware is all new (the old hardware will be converted to a testing machine), but the hard-disk is my original one, containing Redhat 8.0 and all my files. Obviously, the first time I booted the new machine Kudzu ran and removed old hardware's drivers and discovered some new harder (e.g., different network cards, video card, etc.). Almost everying seems to be working on the new machine, except the mouse's wheel stopped working. I know the mouse wheel is an evil Microsoft invention which imitates Tk's middle mouse button bindings, but still, I've gotten used to it when browsing in Mozilla, and I want it back... ;) When I have in /etc/X11/F86Config lines like Option "Device""/dev/mouse" Option "Protocol" "PS/2" Option "ZAxisMapping" "4 5" The mouse works well, but the wheel is ignored. This makes sense, as the protocol should be "IMPS/2", not "PS/2". But when I change it to IMPS/2 and restart X (or even reboot), my mouse goes berserk, and randomly moves about, tending to stick to the lower-left corner and moving at terrific speeds, but only when it feels like it. Has anybody seen this phenomenon before? Needless to say, this is the same wheel-mouse I always had on my old machine (a Logitech M-S48a) and the same Linux and X11 software I had before... Can it be that my new BIOS is somehow screwing with the mouse? Or that the new USB on the system (which I don't plug anything into) somehow got X11 confused with fake mouse events? Or something else? I'll be grateful for any hints.. -- Nadav Har'El| Monday, Mar 17 2003, 13 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |From the Linux getopt(3) manpage: "BUGS: http://nadav.harel.org.il |This manpage is confusing." = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Request for update on the current Linux Distributions.
On Mon, Mar 17, 2003, Yedidyah Bar-David wrote about "Re: Request for update on the current Linux Distributions.": > Hi, > > Is there a real, urgent, need for a new religious war? There are tons > of places on the net for such comparisons. Why don't you search google > for e.g. "linux distribution comparison"? > Last time there was a weeks-long thread about gentoo/Debian/RedHat and > some more, without conclusive results. I do not think this time (a few > months later) we will have some radical changes. I think in this case, there's no reason for a flame war :) The guy basically asked for a "faster distribution". With all due respect to the various distributions and their specific advantages and disadvantages, significant speed difference isn't one of these things. Other that +-10% that you can perhaps gain from better compilation parameters, the performance of all GNU/Linux distributions are basically the same because they all use basically the same components. Even if the speed difference was 20%, it isn't something you'd notice as a desktop user. If you find, say, OpenOffice, to be a huge memory hog and annoyingly slow to start, well, it will probably be just as large and slow on another distribution. If one distribution ever finds a magical way to make OpenOffice smaller and faster, you can bet that the other distributions, or even the OpenOffice creators themselves, will pick up on that trick and pretty soon all distributions will have it. So I suggest you chose your distribution based on the variety of software it includes, the quality of its integration, the support and update system, and your personal habits and tastes. Speed is probably not a major factor in choosing your distribution - unless for your specific needs require every percent of speed that you can possibly get. -- Nadav Har'El| Monday, Mar 17 2003, 14 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |"Mommy! The garbage man is here!" "Well, http://nadav.harel.org.il |tell him we don't want any!"- Groucho Marx = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Request for update on the current Linux Distributions.
On Mon, Mar 17, 2003, [EMAIL PROTECTED] wrote about "Re: Request for update on the current Linux Distributions.": > On Mon, Mar 17, 2003 at 09:21:38PM +0200, Nadav Har'El wrote: > > The guy basically asked for a "faster distribution". With all due respect > > to the various distributions and their specific advantages and disadvantages, > > significant speed difference isn't one of these things. Other that +-10% >... > But your reply just shaked my memory about some distribution a few years > ago which was about compiling the entire Linux (based on Debian, maybe?) > with the most optimization options possible, they used to have quite > faster execution times back then (30%+). Does anyone rememeber what I'm > talking about? What happened to them? Maybe you're referring to "gentoo" that do such a compilation. It's very hard to obtain more than 10%-15% speed increase from gcc just by compiling for a specific processor (say 686) instead of 386. I believe, though, that even if there were a 30% speed increase in desktop applications you wouldn't be able to notice it... By the way, for some applications the run-time itself is not the problem, but the problem is the annoying start-up time, with dozens of shared libraries, or interpreted script files, being loaded on every run. Some applications, like Emacs and TeX, solved this by dumping precompiled code. KDE improved this by allowing "prelinking" - see for example prelink(8) (it was written by a guy from Redhat, but every distribution can add it). Of course, a distribution can also contain statically-linked executables for improved performance - but I've yet to see any of those (other than Embedded Linux distributions). -- Nadav Har'El|Tuesday, Mar 18 2003, 14 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Seen on the back of a dump truck: http://nadav.harel.org.il |<---PASSING SIDE . . . . . SUICIDE---> = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Somebody mickeyed with my mouse?!
On Mon, Mar 17, 2003, Boris Sukholitko wrote about "Re: Somebody mickeyed with my mouse?!": > On Mon, Mar 17, 2003 at 05:42:06PM +0200, Nadav Har'El wrote: > > The mouse works well, but the wheel is ignored. This makes sense, as the > > protocol should be "IMPS/2", not "PS/2". But when I change it to IMPS/2 and > > restart X (or even reboot), my mouse goes berserk, and randomly moves about, > > tending to stick to the lower-left corner and moving at terrific speeds, but > > only when it feels like it. > > I've had the same problem in my Debian box a couple of days > before. There obviously was some kind of conflict with gpm. > I've solved it by brutally removing gpm package from my > system. For me, disabling gpm (I did "chkconfig --del gpm" and killed gpm, not removed the RPM, but this should not make any difference) did not help. Following the advice of some people, I even powered off the machine, took out the plug, waited and then plugged it back in - because perhaps my mouse has the memory of an elephant and remembers for some reason it should be a PS/2. But none of that helped. My mouse still refuses to work correctly when the X configuration contains IMPS/2. I also checked with "od /dev/mouse" that no mouse events are generated when I move the mouse wheel - so apparently my mouse is indeed in some PS/2 mode... Is there a utility, ioctl, or something to change the mode of the mouse? Could it be that the BIOS is forcing my mouse to be PS/2? I looked in the BIOS's setup menus, but couldn't find anything about that... Well, I guess I need to relearn to use the scroll bar... :) -- Nadav Har'El|Tuesday, Mar 18 2003, 14 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Hardware, n.: The parts of a computer http://nadav.harel.org.il |system that can be kicked. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: backup woes
On Thu, Mar 20, 2003, Michael Sternberg wrote about "backup woes": > /dev/hda6: Can't read next inode while scanning inode #2453824 > > After that dump quits without finishing backup. > > The questions are: > 1. What that mean ? Is my hard disk is gone ? Maybe you should run fsck on the disk to check it... You should unmount it first (not trivial if this is your /bin's filesystem). You can use it with the "-c" option to also check the disk for bad blocks (see e2fsck(8)). You might need to use the "-f" option to fsck, if the filesystem is claimed to be "clean" without any checking being done. > 2. How can I check out to what file belongs this inode ? A slow method would be find /whatever/mountpoint -inum 123456 I don't know of a faster way (there might not be, considering that in Unix the same inode can be used by several files, via hard links). -- Nadav Har'El| Thursday, Mar 20 2003, 16 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |"I don't use drugs, my dreams are http://nadav.harel.org.il |frightening enough." -- M. C. Escher = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: XFree86 4.3 and keyboard groups
On Fri, Mar 21, 2003, Tzafrir Cohen wrote about "Re: XFree86 4.3 and keyboard groups": > > after upgrading to XFree86 4.3, I can no longer switch between the groups in > > my keyboard layout, which means that while I'm using the israeli layout I can > You need to replace "il" with "us,il" (or "il,us" if that's what you > really want) in /etc/X11/Xkbmap , ~/.Xkbmap , /etc/X11/Xf86Config-4 or > whatever . Too bad XFree86 are doing these backward-incompatible changes... I also fail to see the reasoning behind this change. Any Israeli keyboard has both Hebrew and Latin (ASCII) characters on it - there's nothing "US" (American) about this keyboard, except the fact that ASCII was indeed invented in America... All Hebrew keyboards I ever saw in Israel where in this so-called "us,il" configuration. I think that the new, latin-character-less, keyboard, should probably be renamed to "he", or at least "he-il", because the name "il" doesn't seem appropriate for it any more. Oh, well... I pitty all the newbies that will have to tackle this problem when upgrading their Hebrew-supporting Linux system. I hope that at least KDE and Gnome will know to set this up correctly when told that the person has an "Israeli" keyboard. -- Nadav Har'El| Thursday, Mar 20 2003, 16 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Why do doctors call what they do http://nadav.harel.org.il |practice? Think about it. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: output from a remote host
On Sun, Mar 23, 2003, Orna Agmon wrote about "output from a remote host": > My problem is that rsh does not work well in the background (in other > words, ^z bg brings it to a state of "suspended (tty output)"), since both > rsh and rexec need the terminal's standard output as thier own. I believe your problem is not with the standard output, but rather with the standard input - rsh tries to pass the standard input to the remote shell, and you cannot do that for a background process. Try the "-n" option of rsh to solve this. -- Nadav Har'El| Sunday, Mar 23 2003, 19 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A: the American Association for the http://nadav.harel.org.il |Abolition of Abbreviations and Acronyms = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: output from a remote host
On Sun, Mar 23, 2003, Nadav Har'El wrote about "Re: output from a remote host": > On Sun, Mar 23, 2003, Orna Agmon wrote about "output from a remote host": > > My problem is that rsh does not work well in the background (in other > > words, ^z bg brings it to a state of "suspended (tty output)"), since both > > rsh and rexec need the terminal's standard output as thier own. > > I believe your problem is not with the standard output, but rather with > the standard input - rsh tries to pass the standard input to the remote > shell, and you cannot do that for a background process. > > Try the "-n" option of rsh to solve this. Oh, and I forgot, where did you expect "rsh"'s output to go after you put it in the background?? You probably wanted to redirect its output somewhere, as in rsh -n anothercomputer "dosomething" >somefile 2>&1 & If you don't expect any output, redirect it to /dev/null: rsh -n anothercomputer "dosomething" >/dev/null 2>&1 & -- Nadav Har'El| Sunday, Mar 23 2003, 19 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If I am not for myself, who will be for http://nadav.harel.org.il |me? If I am only for myself, who am I? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Redhat 9 going out next week???
I was shown today a mail supposedly announcing that Redhat 9.0 will be out next week. I was very suspicious - that email had certain suspicious and spammy characteristics, not to mention I couldn't believe Redhat will skip 8.1 and jump directly to 9.0. That's what I thought, until I, just to be safe, sent my browser to http://www.redhat.com. And lo and behold, what do I see there? "Today: Get Red Hat Linux 9 early" Linking to the following page: http://www.redhat.com/mktg/rh9iso/ Which says Red Hat 9 will be available on March 31, 2003 for "Red Hat Network" subscribers, and generally available in April 7, 2003. Weird. Very weird. Can anybody confirm or deny, or venture guess why the sudden version number jump? It's a bit too early for an April's Fools hoax... -- Nadav Har'El| Monday, Mar 24 2003, 21 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If I were two-faced, would I be wearing http://nadav.harel.org.il |this one? Abraham Lincoln = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ANSI C
On Tue, Mar 25, 2003, Mark Veltzer wrote about "Re: ANSI C": > P.S. Ritchie and Kerninghan didn't even dream of GCC when they thought of C so > the small book is of little use these days. Actually, Kernighan & Ritchie have a second edition of their book, published in 1988, which corresponds to ANSI C. It might not be the most recently published book, but it's far from "of little use these days", and is still very relevant. (if you think 1988 is old, remember that the first edition of their book was published in 1978). And the second edition is less "small" than the first edition. Harbison & Steele's book, "C Reference Manual" is considered by many to be a better book and with more details about the standard. And of course, there is the text of the standard itself, which ISO probably charge $1000 (or something similarly outrageous) but if you're into illegal copying probably can be found and copied in electronic format. -- Nadav Har'El|Tuesday, Mar 25 2003, 21 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Unlike Microsoft, a restaurant will give http://nadav.harel.org.il |me food for free if I find a bug in it! = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ANSI C
On Tue, Mar 25, 2003, Vadim Vygonets wrote about "Re: ANSI C": > > Actually, Kernighan & Ritchie have a second edition of their book, published > > in 1988, which corresponds to ANSI C. > > It's quite hard to get the first edition these days, don't you > agree? Not for me :) Anyway, I didn't understand the comment about their Second Edition being not relevant or not useful. It is useful, but not a very good reference when it comes to standard-conformance issues. > > but if you're into illegal > > copying probably can be found and copied in electronic format. > > A publically available draft from August 3, 1998: Well, I suppose this cannot be a draft of a standard that was defined over 8 years earlier... :) I suppose it is a draft of a newer standard (?) called C9X. When people speak of ANSI C, they don't normally refer to C9X. If you want your programs to be very portable, you better not rely on C9X extensions. -- Nadav Har'El|Tuesday, Mar 25 2003, 21 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If you tell the truth, you don't have to http://nadav.harel.org.il |remember anything. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ANSI C
On Tue, Mar 25, 2003, Vadim Vygonets wrote about "Re: ANSI C": > > I suppose it is a draft of a newer standard (?) called > > C9X. When people speak of ANSI C, they don't normally refer to C9X. If you > > want your programs to be very portable, you better not rely on C9X extensions. > > "ANSI C" is ISO C for more than a decade now, and the latest(?) So what? I tend to give credit to ANSI for working on this, and call it ANSI C. I also use the term ASCII, not ISO 646... > version is C99, slightly over 3 years old now, so I argue that > C99 is the Standard C today, and this reference is more relevant > today than that to the older version of the standard. Also, the How is it more relevant? I would argue the opposite, that most C compilers do not yet fully support C99, but 99% of the C compilers you'll find nowadays are compliant with the 1990 ANSI C. > discussed piece of code doesn't use any C99 "extensions", as you > put it, so theoretically the standard should be the same in both > versions. I guess you're right, and it won't hurt to look at the C99 standard if it's more readily available. -- Nadav Har'El|Tuesday, Mar 25 2003, 21 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Always go to other people's funerals, http://nadav.harel.org.il |otherwise they won't come to yours. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SLL gateway
On Tue, Mar 25, 2003, gili gili wrote about "SLL gateway": > I?m trying to set up a SLL gateway, what I mean is to create one server, > and behind him all my http & https server. The client connect to the ?SLL > gateway? in https, the ?SLL gateway? unwrap the https read the http header, > My questions are: > 1) Is this architecture looks reasonable, or am I fighting windmills??? I don't know if this is possible in squid, I never actually tried to use Squid with SSL, but it is certainly possible to run Apache + Mod_ssl in the mode you describe (if I understood correctly what you described). Another thing you'll need to worry about is that SSL work, especially the server-side RSA, is pretty slow, so unless you get a hardware acceleration card for SSL, the performance of this setup might disappoint you. Several companies also sell integrated devices which do the things you describe, which are called "SSL accelerators", and are probably better in performance, scalability, and security than some setup you'll concoct yourself in an afternoon. One of these companies is Radware (www.radware.com), an Israeli company I work for; Radware's SSL Accelerator is called "CertainT 100". > 3) If any one tried this kind of things (SLL reveres proxy, SLL wrappers, > etc), can U give me some millstones? "Stunnel" is a decent SSL wrapper. It might, or might not, be enough for your needs. P.S. It's "SSL" (Secure Socket Layer), not SLL. -- Nadav Har'El| Wednesday, Mar 26 2003, 22 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I planted some bird seed. A bird came up. http://nadav.harel.org.il |Now I don't know what to feed it... = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SLL gateway
On Thu, Mar 27, 2003, Shachar Shemesh wrote about "Re: SLL gateway": > Out of curiosity - don't you get certificate authentication error when > connecting to the sites, that the name on the certificate doesn't match > the name of the site? Typically, an SSL accelerator (or a cluster of such devices) is used in front of a web-server farm; The SSL accelerator's IP address is the published address of your site, the DNS refers to that IP address, and the site's certificate is installed inside the SSL accelerator (supposedly in a secure manner, perhaps even using a FIPS-compliant SSL card which holds the site's keys in a way they cannot be stolen by software). Various configuration techniques (the simplest of which is bridge mode) allow you to stick an SSL accelerator in front of a running http server farm without even needing to change IP addresses of anything or modify any DNS settings. -- Nadav Har'El| Thursday, Mar 27 2003, 23 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |How do you tell when a pineapple is ready http://nadav.harel.org.il |to eat? It picks up its knife and fork = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SLL gateway
On Thu, Mar 27, 2003, Shachar Shemesh wrote about "Re: SLL gateway": > Nadav Har'El wrote: > > >the site's > >certificate is installed inside the SSL accelerator > > > > > But that's precisely the problem. The setup you describe is only > applicable when a single SSL proxy handles a single web server. If one > proxy handles several servers, it needs a different certificate for the > different servers. Right. And why is that a problem? An SSL accelrator or a cluster of such devices could potentially do SSL encryption for hundreds of different sites, provided that this number of keys and certificates are put on the device. Obviously, you'll also need an IP address per SSL site you plan to serve (this is a basic SSL limitation - you can't normally do "virtual hosts" with SSL). If you want to read more about what people expect from a full-featured SSL- accelerator, check http://www.radware.com/content/products/ct100/default.asp > between the various servers. If everyone are given the same IP, as is > Gili's case, that cannot be done. If that is not done, it is not > possible to distinguish between the various servers, and SSL server > authentication is not possible. Why is everyone given the same IP? Nothing prevents you him from giving 100 different IP addresses to one computer... -- Nadav Har'El| Thursday, Mar 27 2003, 23 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A Nobel Peace Prize? I would KILL for one http://nadav.harel.org.il |of those. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: go-linux date?
On Thu, Mar 27, 2003, Dan Armak wrote about "go-linux date?": > I thought it was April 10th as stated here, but www.pc.co.il/go-linux/ now > says Apr. 8th. Has it been moved? Quoting that page: "מועד: 10/4/03 מקום: מלון הילטון, ת''א - חדר המליאות" So I don't see any evidence of this date change... Where did you see 8/4? -- Nadav Har'El| Friday, Mar 28 2003, 24 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Help Wanted: Telepath. You know where to http://nadav.harel.org.il |apply. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux for machines with large memory
On Fri, Mar 28, 2003, Yedidyah Bar-David wrote about "Re: Linux for machines with large memory": > While I did not try that myself yet (but did use up to 4GB), from what > I read it's trivial - simply select 64GB in "High Memory support" > when you compile your kernel (CONFIG_HIGHMEM64G=y). And if you have between ~0.9 to 4 GB, you'll need to select "4GB" in that menu. PAE does not have to be used, but other tricks do, to be able to map virtual memory and (most, or some of) physical memory into the same address space. > Each process will still be limited to 4GB. Actually, it will be limited to slighty less than 3GB, because the remaining 1GB is used by the kernel, by default, to map physical memory. With some hacking you can get this limit closer to 4GB but at the price of the kernel having less physical memory to use for DMA and stuff like that. As far as I know (and I'd be very happy to be corrected about that!) there is no way to have one process on an i386 be able to use more than 4GB of virtual memory; That's because this process will be using pointers, and i386 points are 32bit wide and so can only refer to 4GB of address space. Anyway, as I said in the previous paragraph, your practical limit for one process will be (by default) 3GB, not 4GB. -- Nadav Har'El| Sunday, Mar 30 2003, 26 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If I am not for myself, who will be for http://nadav.harel.org.il |me? If I am only for myself, who am I? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: openoffice 1.1beta is out
On Sun, Mar 30, 2003, Shoshannah Forbes wrote about "Re: openoffice 1.1beta is out": > OpenOffice 1.1. Does not Have Hebrew support. Only the > developer builds from 643 and up have it. Strange... On openoffice.org (http://news.com.com/2100-1046-994264.html?tag=fd_top) I read: "Significant additions in version 1.1 include ... enhanced support creating non-English versions of the software, including languages that read text right-to-left or bottom-to-top." and in http://www.openoffice.org/ I read "OpenOffice.org 1.1beta is ready for immediate download. ... Support for Complex Text Layout (CTL) and languages such as Thai, Hindi, Arabic, Hebrew" So either Hebrew support is indeed inside this beta, or they are just referring to some theoretical "support" that hasn't actually been compiled in. If the latter, why?? It's not like they want to keep OpenOffice small :) ) -- Nadav Har'El| Sunday, Mar 30 2003, 26 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A bird in the hand is safer than one http://nadav.harel.org.il |overhead. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: openoffice 1.1beta is out
On Sun, Mar 30, 2003, Barak Bloch wrote about "Re: openoffice 1.1beta is out": > Hebrew work very well. > I am using it right now, and its much better than the 643c i had. > And its work for me with hebrew out-ot-the-box. Does Hebrew also work in the presentation creation application in openoffice? If so, I'll consider making my go-linux presentation using OpenOffice :) > And after using "export to PDF": > http://forums.ort.org.il/files/117/1729028/2032407.pdf This looks quite bad for me, with punctuation going over the Hebrew text (I tried acroread, ghostview and xpdf, with the same problem). Is there an option to create better-looking postscript output? -- Nadav Har'El| Sunday, Mar 30 2003, 26 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Computers are like air conditioners. Both http://nadav.harel.org.il |stop working if you open windows. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: permanent IP address
On Sun, Mar 30, 2003, shlomo solomon wrote about "permanent IP address": > Here are the symptoms I had: > 1 - I connected and could ping, but packet loss was about 70% When ping'ing, did you use the -n option? Otherwise DNS problems might still be causing your problems. > 2 - When pinging a URL like www.iglu.org.il (and not a numerical IP address) I > got no reaction at all (possible DNS problem, but I had no DNS problems with > the dynamic address) Did you set up /etc/resolv.conf yourself, or were you hoping that the dialin process would do that for you (maybe it doesn't)? Could it be that you are using a name server but it refuses to serve your new IP address, for some reason? Try "nslookup" manually to see what happens when you do DNS searches. Sorry, I don't have a better guess on what is happening on your system. Good luck. -- Nadav Har'El| Sunday, Mar 30 2003, 27 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |The person who knows how to laugh at http://nadav.harel.org.il |himself will never cease to be amused. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Well ... it happened again !
On Mon, Mar 31, 2003, Eli Segal wrote about "Well ... it happened again !": > Yesterday my computer has stuck on me again ! Yeah, that's what happens when pranksters pour glue on your keyboard :) > and .. nothing , i have to press my computer button > even SysRq doesn't work Does Ctrl-Alt-F1 (to go to another virtual terminal) work? Can you ping your computer from another computer? Does the mouse cursor move? Which distribution are you using? What kernel? Any other special drivers or devices on your system? We have way too little information to help you other than making wild guesses, sorry... -- Nadav Har'El| Monday, Mar 31 2003, 27 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Cats are smarter than dogs. You can't get http://nadav.harel.org.il |eight cats to pull a sled through snow. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Spammed by my ISP (Bezeq Int.)
On Wed, Apr 02, 2003, Arik Baratz wrote about "Spammed by my ISP (Bezeq Int.)": > It is clear that the address this message was sent to, [EMAIL PROTECTED], was > harvested from the ISOC database, in contradiction with their terms and conditions. > I have never given this address in any other opportunity. > > My question is: If you complain to the ISP when you are spammed by someone in their > address range, who do you complain to when the ISP spams you? Is there a higher body > which controls what the ISPs in Israel can or can't do? Is the ISP's upstream > provider the only avenue for my complaint? You said your self that the ISP abused ISOC's list, so you can try to complain is ISOC. And maybe once and for all these registrars will stop forcing you to make all your details public (including email address, home address, and phone) just because you want to have a domain name... And for everybody saying "but these details helps us catch spammers and the likes", well, this is far from being true. The real abusers can easily provide false information, and I've seen false information even on reputable companies' assignments. Besides, if someone chooses not to provide this information he should be allowed to, but it will just cause a easier finger on the black-list trigger (just like happens now regarding to Chinese domains), so that serious non- spammer hosts will volunterily provide this information. Also, if you're a customer of that ISP, can't you complain to them directly, saying you'll be taking your business elsewhere? -- Nadav Har'El| Wednesday, Apr 2 2003, 1 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |This signature was intentionally left http://nadav.harel.org.il |boring. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Spammed by my ISP (Bezeq Int.)
On Thu, Apr 03, 2003, Vadim Vygonets wrote about "Re: Spammed by my ISP (Bezeq Int.)": > These details help people report problems when other ways of > contact are unavailable for some reason (network downtime, > misconfigurations, etc.). I used RIPE's whois database to get > phone numbers of domain holders whose mail servers were > misconfigured. All I said was that this information should be voluntery. If a domain holder does not want to be helped when his name server is down, it should be his choice. Imagine your busy domain going down, and suddenly 100 people start calling you to tell you about it, while all the time you *know* of this problem and are busy fixing it. Administrators should be given the choice of whether they want such phone calls or not. And this is doubly true for mail address - I can't think of any legitimate reason I would want somebody to come to my home because I hold a certain domain... -- Nadav Har'El| Thursday, Apr 3 2003, 1 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I am the world's greatest authority on my http://nadav.harel.org.il |own opinion. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Lyx 1.3
On Fri, Apr 04, 2003, Dekel Tsur wrote about "Re: Lyx 1.3": > BTW, what is the difference between LANG and LC_ALL ? Take a look at the setlocale(3) man page - it explains all these variables. You'll see there that the difference between LANG and LC_ALL is their "priority": LC_ALL overrides any other LC_* variable, while LANG is used as the default for any LC_* variable which is not defined. -- Nadav Har'El| Saturday, Apr 5 2003, 3 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Unix is user friendly - it's just picky http://nadav.harel.org.il |about it's friends. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: In what ways maildir is probably better then mbox?
On Sun, Apr 06, 2003, Tzafrir Cohen wrote about "Re: In what ways maildir is probably better then mbox?": > Think also of delivery of mail to such a mailbox: delivering a mail to a > maildir is certainly much faster. So is delition. For deletion your point is obviously true, but I don't think it's true for delivery: delivery in an old-style mailbox file involves simply appending a message to the end of the message - there is no need to search the big mailbox for anything while doing it. -- Nadav Har'El| Sunday, Apr 6 2003, 4 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Always go to other people's funerals, http://nadav.harel.org.il |otherwise they won't come to yours. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: In what ways maildir is probably better then mbox?
On Sun, Apr 06, 2003, Ira Abramov wrote about "Re: In what ways maildir is probably better then mbox?": > > For deletion your point is obviously true, but I don't think it's true for > > delivery: delivery in an old-style mailbox file involves simply appending > > a message to the end of the message - there is no need to search the big > > mailbox for anything while doing it. > > other than seeking to the very end of it, locking it, checking for > previous locks, checking for stale status of said locks, etc. etc. Everything you describe is an O(1) operation, not something that becomes longer as the mailbox grows. The small O(1) operations you describe are not noticable in any situation except perhaps when you are serving hundreds of thousands of mailboxes. By the way, maildir implictly includes the same locks and everything you describe, it's just that they are inside the kernel (so that two processes writing to the same directory concurrently don't ruin it). > think of maildir as a hash and mbox as a linked list. it's even worse As I said, it is only the *append* operation which is fast on mbox - other operations like showing the list of messages or deleting a message in the middle of the file indeed require reading or writing the whole file which is very inefficient for large files. > since you don't know how large each message is when you traverse the > mbox so you need to read each and every byte into memory, think about > cache thrashing and trashing with modern mailboxes full of large > attachments. This is not entirely true, because modern mail messages contain a Content-Length: header allowing to skip large emails and attachments easily. The more problematic case is that of an mbox containing thousands of small messages. By the way, even with the mailbox file format, a smart mail browser could optimize the "perceived response time" by, say, showing the list of the last 500K of messages first (supposedly, the newest and most interesting to the user), and then going back to showing the rest of the messages - just like web browsers nowadays display web-pages incrementally, as they are being loaded. > and a word about "indexed mailboxes", Mozilla doesn't change a thing in > the mbox format, it just adds a persistant index file that saves the > time of indexing an mbox in RAM each time you open it. If you dump the > index you should still be able to read it in mutt for instance. some > past versions of Netscape knew how to reindex standalone mbox files when > forced to adopt one, I have no idea if that option still exists. Another possible optimization (that I don't know if Mozilla did or did not) is that there is no need to rewrite the mbox every time you delete a few messages - it is enough to mark messages deleted, and do the "gap closing" and reindexing later - perhaps even as a cron job when the user isn't waiting on this folder. -- Nadav Har'El| Sunday, Apr 6 2003, 4 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |This box was intentionally left blank. http://nadav.harel.org.il | = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: In what ways maildir is probably better then mbox?
On Sun, Apr 06, 2003, Oleg Goldshmidt wrote about "Re: In what ways maildir is probably better then mbox?": > Another problem that is often mentioned is that a lot of inodes are > used up. I don't know if reiser FS solves that problem, too. Supposedly, it does. Quoting www.namesys.com: "We don't have fixed space allocation for inodes. That saves 6% of your disk. ... In my approach I store both files and filenames in a balanced tree, with small files, directory entries, inodes, and the tail ends of large files all being more efficiently packed as a result of relaxing the requirements of block alignment, and eliminating the use of a fixed space allocation for inodes. " -- Nadav Har'El| Sunday, Apr 6 2003, 4 Nisan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Anyone who quotes me in their sig is an http://nadav.harel.org.il |idiot. -- Rusty Russell's sig. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SCO SCO SCO (or whatever)
On Sun, Jun 01, 2003, Yehuda Drori wrote about "Re: SCO SCO SCO (or whatever)": > Shahar and all other on linux-il whom concern. Shachar already replied to you well, explaining why he didn't advertise on whatsup because it wasn't on *his* list of most-frequently visited sites, but that shouldn't stop whatsup from covering Hamakor activities and announcements, or from virtually anybody (including you yourself) from picking up Hamakor announcements from other sources and sending them to whatsup. Seeing whatsup's look and feel, I assume that whatsup wants to be an Israeli slashdot-like site, right? Well, as you can see slashdot is not a site in which *original* news is presented - it *always* points to news written in other sites (it just organizes it better and picks the most interesting pieces of news). So you should not be offended if Hamakor news appears on other sites (hamakor.org.il, etc.) and whatsup just points to it... And on a different topic, here are a few quotes from your message: > whatsup is defiantly today the largest place on then net which deals > with LOS matters and try to bring as much info and knowledge to the public. >.. > many of the sec. group I mentioned above visiting on a regular daily basis >... > the best and widest channel he can get to it.. >... > get out of the bubble the linux-il and Iglu and get more involved in > the forming community that takes place in the last months. >... Ok, care to back up your claims? You claim that whatsup is the biggest and best discussion forum in Israel about Linux, and that linux-il is a bubble filled with elitistic elders. How many actual readers (not one-time visitors) does whatsup have (linux-il has several hundreds, I think)? How many threads on whatsup get meaningful replies and not just "chatter"? How many of the "really important" announcments get sent to there? How many of the "second group" (newbies) actually read whatsup every day like you said? What makes whatsup the *best*, like you said? Don't get me wrong, I do think that whatsup is an interesting site, I do read it quite often and I even participated in a few threads (as an "anonymous linuxer"). But I think that some of your claims about its importance are a bit exaggerated, and the thought that Hamakor (or Shachar in this case) failed to contact Globes just because they didn't write on whatsup is really silly, in my opinion. The sad truth (or at least, my guess) is that any article in globes or article on the online ynet probably get read by 10-100 times the number of people that read whatsup. If something appears in the printed Yediot Achronot, the ratio probably jumps to 1000. So Shachar *did* try to contact what he probably considered the best and most-read media! P.S. I'm not an Hamakor representative in any way, form or shape, so please don't go trashing Hamakor based on what I wrote above... -- Nadav Har'El| Sunday, Jun 1 2003, 1 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |I before E except after C. We live in a http://nadav.harel.org.il |weird society! = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SCO SCO SCO (or whatever)
On Tue, Jun 03, 2003, Shachar Shemesh wrote about "Re: SCO SCO SCO (or whatever)": > Creating awarness with those journalists take time. We sent one press > release about Haifux already. I missed the HSpell announcement. Since Hspell became a favorite example here, I think I better comment on it. But keep in mind that Hspell has *NOTHING* to do with Hamakor, and our decisions regarding it are completely separate from Hamakor's board's decisions regarding their activities. Every time we (Dan Kenigsberg and I) release a new version of Hspell, I send an announcement to exactly three places: 1. linux-il (this list) 2. ivrix-discuss (which unlike this list, isn't drowned with offtopic discussions like this one). 3. freshmeat.net I send it to these sources because, frankly, these are *my* favorite sources of information. I don't send further announcements to ynet, globes, hamakor, whatsup, penguin, slashdot, debian, tapuz, or one of the dozen other forums which might be interested in this announcements, which doesn't mean that all those projects and others aren't worthy and wonderful sources of information, perhaps even better than the three that I chose and perhaps with a larger crowd. I realize that it would be more convenient to Whatsup if I send them an announcement in Hebrew, to debian if I send them a tested .deb, and so on, but unfortunately my time is limited. Next time, I promise I'll think about announcing on Whatsup too. But in the meantime, if people find Hspell valuable enough (and fortunately, they seem to do) they will come forward and do those things themselves - like an Hspell announcement appeared on whatsup even if I didn't write it, or .deb packages appeared (thanks to Baruch Even) even though we didn't know anything about how to create them. Anyway, keep in mind that Hspell is a technical project, not a PR fad; For Hspell to succeed, it is frankly more important for it to have technical merit (good programming, good linguistic data) than good PR. When competing against a void (i.e., the zero other free Hebrew spell-checkers), good PR isn't a deciding factor - though it certainly never hurts. Getting Hspell coverage in websites and newspapers is flattering, but is not our goal: our goal is to make hspell high-quality enough, and get it into popular distributions; If Hspell was installed with Redhat, for example, we wouldn't need any PR for new Hspell versions, just like 99.9% of the "ls" users never read announcements of new versions of it. -- Nadav Har'El| Wednesday, Jun 4 2003, 4 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Entropy: Not just a fad, it's the future! http://nadav.harel.org.il | = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: C flame (was: FS/OS in schools)
On Fri, Jun 06, 2003, Herouth Maoz wrote about "Re: C flame (was: FS/OS in schools)": > Quoting Nadav Har'El <[EMAIL PROTECTED]>: > > > Let's face it, when you teach someone to drive its enough to teach them > > to use the pedals and the steering-wheel. But when you teach someone to > > become a mechanic, you have no option but showing them what the engine > > looks like by opening the hood, even if it's an ugly, greasy, mess. > > But you don't teach somebody to be a mechanic before he has learned what a car > is. I don't think there are many car repairsmen who do not have a drivers' > license, or at least a scooter license... To continue this silly metaphore, *driving* a car corresponds to using a computer for stuff needed for daily life, say, playing or writing papers for school. For that you needn't learn any programming at all. And indeed, like you said, most people *use* a computer before they start learning to program on it. But if you do want to learn how to fix your own car (or to program your own computer) I argued that you better understand the nuts and bolts before you understand the fancy new fuel-injection system ;) -- Nadav Har'El| Saturday, Jun 7 2003, 8 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |If God is watching us, the least we can http://nadav.harel.org.il |do is be entertaining. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Redhat 9 going out next week???
On Mon, Mar 24, 2003, Hetz Ben Hamo wrote about "Re: Redhat 9 going out next week???": > > Which says Red Hat 9 will be available on March 31, 2003 for "Red Hat > > Network" subscribers, and generally available in April 7, 2003. > > > > Weird. Very weird. Can anybody confirm or deny, or venture guess why the > > sudden version number jump? > > The jump was due to the incompatibilities rolled - like the famous glibc which > is NOT backward compatible, etc... It's a strange explanation, because Redhat just sent out the new glibc as an errata for Redhat 8.0 (because of some vulnerability discovered in the XDR code of the old glibc). So the new glibc is ok as an errata (that many people install without even thinking twice) but not as Redhat 8.1? trange... Maybe the real explanation is that they are trying to beat Mandrake's version number ;) -- Nadav Har'El| Monday, Mar 24 2003, 21 Adar II 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |An egotist is a person of low taste, more http://nadav.harel.org.il |interested in himself than in me. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: FS/OS in schools: why don't *they* tell us what they want?
On Sun, Jun 08, 2003, Stanislav Malyshev wrote about "Re: FS/OS in schools: why don't *they* tell us what they want?": > Government decision that all schools should use OpenOffice would also cost > the taxpayers (maybe less) and lock out the competition. The only visible > difference is that "we" win and "they" lose. That's OK, but speaking in > these terms and speaking about public benefit and freedom in the same time > doesn't sit well. Wrong. If OpenOffice "won", and everyone started sending OpenOffice files to one another, Microsoft (or any other word-processor manufacturer) could relatively-easily use OpenOffice files because the format is open and code that reads and write it is available. The code is probably GPL so a proprietary program could not use it directly, but since it not patented nothing prevents programmers from reading the code and writing other code that does the exact same thing. So OpenOffice could never have the "edge" that Microsoft now has - being the only software with 100% compatibility with a commonly-sent file format - and it could not "lock out the competition". In any case, I'd prefer it that some open format (like OO's) "won" but the market of actual word-processing programs remain divided with no clear winner. This will ensure that nobody will *assume* that you must use a certain kind of word processor, and everyone will be free to choose his or her own according to its price, its features, its bloat (or lack thereof), its support of Hebrew, or whatever. -- Nadav Har'El| Monday, Jun 9 2003, 9 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Live as if you were to die tomorrow, http://nadav.harel.org.il |learn as if you were to live forever. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
new article in ynet
Gal Mor from ynet continues to bring the gospel of free software to the masses (or at least, to ynet readers), with his new article, http://www.ynet.co.il/home/0,7340,L-745-2653760,00.html titled "On Piracy and Open Source". He described the fallacy of statements like "13 billion dollars were lost last year due to piracy", and explains how for many users, such as business users in developing countries, buying expensive software is not even an option, and not in the best interest of their country; He then continues to explain why free software is an option for these people - a good option. P.S. I tried announcing this in Whatsup, but entering bidi (hebrew text plus links) on the site was all but impossible (and very confusing!), so I gave up :( -- Nadav Har'El|Wednesday, Jun 11 2003, 12 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |echo '[q]sa[ln0=aln256%Pln256/snlbx] http://nadav.harel.org.il |sb3135071790101768542287578439snlbxq'|dc = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: What programming language to teach in schools ?
On Wed, Jun 11, 2003, Beni Cherniavsky wrote about "Re: What programming language to
teach in schools ?":
> I also think something like the abstract equivallent of UNIX pipes
> should be taught before any langauge at all. The usefulness of pascal
> programs doable in school lessons/exams is a joke. What's the point,
> if you acn't see the bigger picture?
I agree with you here 100%.
Kids should not be taught to program before they understand what programs
are supposed to do. Unix pipes and shell and so on are indeed a good
system to see before starting to program, but understanding how your favorite
system (even if limited like Windows) works is also good.
When I started learning programming (18 years ago), it was with Kernighan &
Pike's "The Unix Programming Environment", which had just come out a year
earlier (1984). This book teaches the way you'll like: it [1] starts with
how to run existing programs (command line, of course, since this was 1984)
and gives examples of the programs available on a Unix system, continues on
why automation is useful (that's why we program, after all!) and how to
automate stuff using pipes and shell scripts. It then continues with more
sophisticated scripting using Awk, sed, and other simple languages, and
then goes to lower-level programming in C, and concludes with a complete
example of how one would write his own language, using Yacc. All of these
steps are acompanied by interesting working examples; In fact some of these
example programs were so interesting that I still use improved versions of
several of their examples till this very day, 18 years later.
It is not a thick book, but it took me almost two years to fully learn
it (at the age of 10, and when I barely knew English, so it wasn't easy :)).
I found this book's approach to be the perfect introduction to programming
and to computers in general.
Too bad that Kernighan and Pike never got back to write a second edition
of their masterpiece, because as it stands it is hard for me to recommend
the 1984 edition to new users; Some of the things they wrote about were
already obsolete in 1985...
And no, learning C first (well, not first: after shell and awk) didn't
"ruin me" in regards to OO languages: I learned C++ about 5 years later
(when I got my hands on the first edition of Stroustrup's book) and I
believe I understand it (and OO methodology) very well. But because I
have a bigger perspective of several programming methodologies, I haven't
become a religious-OO-devotee like some people who learn C++ first have
become.
[1] I'm saying this from memory - sorry if I don't remember it correctly.
--
Nadav Har'El|Wednesday, Jun 11 2003, 12 Sivan 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |:(){ :|:&};: # DANGER: DO NOT run this,
http://nadav.harel.org.il |unless you REALLY know what you're doing!
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: What programming language to teach in schools ?
On Thu, Jun 12, 2003, Daniel Vainsencher wrote about "Re: What programming language to teach in schools ?": > > But because I > > have a bigger perspective of several programming methodologies, I haven't > > become a religious-OO-devotee like some people who learn C++ first have > > become. > That's hilarious. How someone could possibly become an OO fanatic from > working in C++ I have no idea... two qoutes for perspective: Have you never seen people who can't write a 100 line program without insisting using namespaces and classes? Who can't think of any data structure that isn't composed of elements they read about in the "design pattern" books? Who can't discuss an idea without drawing weird diagrams (aka UML)? Who can't imagine ever programming in C because "it's not OO"? Remember the people who used to chant "GOTO is bad! GOTO is bad!" after learning structured programming? These people now start with C++ and learn to chant "global variables are bad", "gloal functions are bad", "functions longer than 10 lines are bad", and these sort of things. -- Nadav Har'El| Thursday, Jun 12 2003, 12 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |A: the American Association for the http://nadav.harel.org.il |Abolition of Abbreviations and Acronyms = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: AOL doesn't accept mail - free relaying of email
On Thu, Jun 12, 2003, Arie Folger wrote about "AOL doesn't accept mail - free relaying of email": > to the outside world, definitely not as a mail server. So, am I right to > conclude that this complaint of AOL's is directed towards my ISP? Or is it > towards the ADSL router of ours? We're missing a lot of information. How do you normally send email? Do you send all the outgoing mail to your ISP's mail server, or do you send mail directly from the ADSL-connected machine? If you use your ISP's mail server for outgoing mail, what server is this? Looking at your post, I see you sent it through 212.40.5.186, the mail server of some Swiss ISP (?!). This mail server is not on any blacklist that I know of, so I don't understand where your problem comes from. So to try to help you, I'd need to see AOL's bounce, and an example of the full headers of your outgoing mail. > Also, I mailed them a complaint letter why they shouldn't maintain such a > policy, but they didn't even bother to reply. For people like me, who get 5-10 times more spam than important email in a day, not dealing with spam is not an option. Certainly it would have been better if every user had the option of how to deal with spam instead of the ISP defining ISP-wide blocks, but frankly AOL's users have the right to switch ISP if the blocking bothers them. -- Nadav Har'El| Thursday, Jun 12 2003, 12 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Drink varnish and you'll get a lovely http://nadav.harel.org.il |finish. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Testing on various computers needed
On Fri, Jun 13, 2003, Ilya Konstantinov wrote about "Re: Testing on various computers
needed":
> 2. time dd if=/dev/zero of=foo
> says:
> real0m10.449s
> user0m0.260s
> sys 0m4.080s
> How come real ("wall clock") time is so much higher than user+sys combined?
This is simple - the disk can't write as fast as the CPU reads from /dev/zero
(i.e., generates a stream of zeros). So half the time, the CPU just waits
for the disk to finish to read. (remember, write buffers, however big they
are allowed to grow, still have a limited finite size).
> 3. nice -20 dd if=/dev/zero of=foo
> doesn't make it any more responsive!
>
> 4. ps shows only ~50% CPU usage.
> There's plenty of CPU time available for the rest.
The 50% CPU average doesn't take into account:
1. The possibility of having alternating seconds of inactivity and 100%
CPU utilization by the kernel (with no chance of any user process running).
If this happens, it's really annoying (especially if you're listening to
background music and it skips).
2. The possibility that no new data can be read or written from the disk
because it's busy. I don't know how good Linux is at reordering disk
requests, but even if it's great at it (and I doubt it) it's easy for
user commands (like "ls") to want to read or write and have to wait for
at least the current sector (and probably much, much, more) to be written
to disk.
I believe that a kernel *can* be designed to reorder disk requests, and
fairly throttle disk requests from multiple processes; Probably Linux as
it exists is simply not very good at this... But I don't really know.
> 5. Only operations which require reads from the hard drive (in my test, I
> write 'foo' to the system drive) cause the slowness. Other operations are as
> smooth as usual.
Which confirms my hunch that crappy request reordering is the root of this
problem.
--
Nadav Har'El| Friday, Jun 13 2003, 13 Sivan 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Preserve wildlife -- pickle a squirrel
http://nadav.harel.org.il |today!
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Command line limit for an arbitrary program?
On Fri, Jun 13, 2003, Shaul Karl wrote about "Command line limit for an arbitrary program?": > 1. What is the limit for strlen(argv[1]) and where is it set? On Unix, command line arguments are passed from the running program to the executable which is going to replace it, with an execve() system call (or one of its variants). These arguments are saved in a special location (later to be given to the C program as a "argv" array) and their total length (not the length of each individual argument) is limited by the kernel. Actually, it's slightly more complicated: the environment-variables are also passed in the same way (and given to C programs as a third "envp" variable that very few people are aware of), and this also takes part of the limited size (if I remember correctly). Anyway, if you look at the execve(2) manual, you'll see that execve() (or the other exec variants) will fail with E2BIG error if the argument list is too long. What exactly the limit is isn't specified in the manual, so you need to either experiment (easy! see if you can write a simple C program to check it out) or to look at the kernel sources to figure out this limit. >From a quick glance on the kernel include files (unfortunately I'm not an expert enough on the Linux kernel to give you an authoritative answer) I see that 32 pages are allocated for arguments. At 4K per page, that comes out to a limit of 131072 bytes for the arguments + environment. Exercise: verify this limit and report back to the group! > 2. What will happen in case it is passed a longer argument? Assuming >root permissions, will it allow an exploit? exec*() will fail and return E2BIG. I've never heard of any way to exploit such a problem, but if a program tries to run another problem and doesn't check for exec() errors and goes on doing weird things, I guess anything is possible... I recommend you always follow exec() with _exit() ! > 3. Is there a compilation or another constant for this length? Maybe modifying binfmts.h (MAX_ARG_PAGES) or limits.h (ARG_MAX) in the kernel sources and recompiling the kernel will change this limit. Exercise: verify this by trying, or by looking at where these constants are actually used inside the kernel. -- Nadav Har'El| Friday, Jun 13 2003, 13 Sivan 5763 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Only dead fish go with the flow. http://nadav.harel.org.il | = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Testing on various computers needed
On Fri, Jun 13, 2003, Nadav Har'El wrote about "Re: Testing on various computers
needed":
> On Fri, Jun 13, 2003, Ilya Konstantinov wrote about "Re: Testing on various
> computers needed":
> > How come real ("wall clock") time is so much higher than user+sys combined?
>
> This is simple - the disk can't write as fast as the CPU reads from /dev/zero
> (i.e., generates a stream of zeros). So half the time, the CPU just waits
> for the disk to finish to read. (remember, write buffers, however big they
> are allowed to grow, still have a limited finite size).
Oops, I meant, of course, "finish to write", not read.
--
Nadav Har'El| Friday, Jun 13 2003, 13 Sivan 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |If you drink, don't park; Accidents cause
http://nadav.harel.org.il |people.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
