There are 21 messages in this issue.
Topics in this digest:
1a. Re: TECH: info on ftp
From: Sai Emrys
1b. OFFLIST Re: TECH: info on ftp
From: David J. Peterson
1c. Re: TECH: info on ftp
From: David J. Peterson
1d. Re: OFFLIST Re: TECH: info on ftp
From: Mark J. Reed
1e. Re: Not OFFLIST Re: TECH: info on ftp
From: Lars Finsen
1f. Re: Not OFFLIST Re: TECH: info on ftp
From: Mark J. Reed
1g. Re: Not OFFLIST Re: TECH: info on ftp
From: Lars Finsen
1h. Re: Not OFFLIST Re: TECH: info on ftp
From: Keith Gaughan
1i. Re: OFFLIST Re: TECH: info on ftp
From: Sai Emrys
1j. Re: Not OFFLIST Re: TECH: info on ftp
From: Sai Emrys
1k. Re: Not OFFLIST Re: TECH: info on ftp
From: Sai Emrys
2a. Re: Another Sketch: Palno
From: Veoler
2b. Re: Another Sketch: Palno
From: Veoler
2c. Re: Another Sketch: Palno
From: Veoler
2d. Re: Another Sketch: Palno
From: Logan Kearsley
2e. Re: Another Sketch: Palno
From: Jim Henry
3a. Average life of a conlang
From: Jim Henry
3b. Re: Average life of a conlang
From: Herman Miller
4a. Playing with dialect variations
From: Logan Kearsley
4b. Re: Playing with dialect variations
From: Philip Newton
5a. Re: Language Sketch: Gogido
From: Philip Newton
Messages
________________________________________________________________________
1a. Re: TECH: info on ftp
Posted by: "Sai Emrys" [EMAIL PROTECTED]
Date: Tue Aug 26, 2008 11:54 pm ((PDT))
FWIW, some services:
http://yousendit.com will store your <100MB attachment for sending in emails
for about a week
http://megaupload.com and its ilk will store yet larger files, but is more a
pain to use
ServU and FileZilla are quite nice and easily configured FTP servers for
Windows, which I've used before. ("Before" here meaning "back in 2000 when I
actually used Windows on a regular basis"...)
I'm not familiar with good Mac software for this, and for *nix the only
thing I've ever used is ssh/scp rather than ftp.
One thing to mind is that FTP, unless you're doing something advanced, sends
passwords and content in the clear over the internet. This means that anyone
between the two computers involved can read any of that data, including the
username/password. So don't use it for anything sensitive, and never ever
use the same password for FTP as you use for anything else... 'cause the
first thing someone malicious would try would be to reuse that same
user/pass on other sites (e.g. web mail services).
All three are capable of doing Samba shares (on Windows it's just called
"[windows] file&folder sharing", on *nix it's the smb:// protocol, and on
Mac it is too though they mostly try to hide it). This is a relatively
convenient method to share things when you can get the two computers
connected. This means though that either you have to allow access from the
outside and have a publicly accessible IP (or forwarded port on the router),
hopefully password protected, or have the computers involved on the same LAN
(e.g. a cable directly connecting them). Samba is at least moderately
encrypted, though it is fairly easy to crack for someone with a modicum of
skill.
In any case, it's a viable alternate method.
FWIW what I personally use: YouSendIt (for <100MB to normal people), one of
the various servers I have access to (scp up, simple HTTP download down), or
direct-cable Samba (for largish transfers between computers that are within
sight of each other). YMMV and all that.
- Sai
Messages in this topic (17)
________________________________________________________________________
1b. OFFLIST Re: TECH: info on ftp
Posted by: "David J. Peterson" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 1:21 am ((PDT))
Hey Sai,
<<
One thing to mind is that FTP, unless you're doing something
advanced, sends
passwords and content in the clear over the internet. This means that
anyone
between the two computers involved can read any of that data,
including the
username/password. So don't use it for anything sensitive, and never
ever
use the same password for FTP as you use for anything else... 'cause the
first thing someone malicious would try would be to reuse that same
user/pass on other sites (e.g. web mail services).
>>
So, I read this, and immediately thought: WHAT?! So, take
my personal website:
http://dedalvs.free.fr/
I use an FTP program to update (upload stuff, etc.). (The program
is called Transmit, by the way. It's for Mac and wonderful, but
it costs money to register [but if you don't register it, you can
still use it for 10 minutes at a time].) But, what exactly does this
mean? Does this mean, essentially, that anyone that navigates
to my site or downloads something off it can access my password
and username? Or is this a different type of FTP transfer that
you're talking about?
-David
*******************************************************************
"A male love inevivi i'ala'i oku i ue pokulu'ume o heki a."
"No eternal reward will forgive us now for wasting the dawn."
-Jim Morrison
http://dedalvs.free.fr/
Messages in this topic (17)
________________________________________________________________________
1c. Re: TECH: info on ftp
Posted by: "David J. Peterson" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 2:44 am ((PDT))
Oh, brother, I sent my reply to the list... I apologize. It's not
that the reply wouldn't be generally useful, but it's not relevant
to conlanging, so I wanted it go offlist. Sorry. :(
-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."
-Jim Morrison
http://dedalvs.free.fr/
On Aug 26, 2008, at 11°53 PM, Sai Emrys wrote:
> FWIW, some services:
>
> http://yousendit.com will store your <100MB attachment for sending
> in emails
> for about a week
> http://megaupload.com and its ilk will store yet larger files, but
> is more a
> pain to use
>
> ServU and FileZilla are quite nice and easily configured FTP
> servers for
> Windows, which I've used before. ("Before" here meaning "back in
> 2000 when I
> actually used Windows on a regular basis"...)
>
> I'm not familiar with good Mac software for this, and for *nix the
> only
> thing I've ever used is ssh/scp rather than ftp.
>
> One thing to mind is that FTP, unless you're doing something
> advanced, sends
> passwords and content in the clear over the internet. This means
> that anyone
> between the two computers involved can read any of that data,
> including the
> username/password. So don't use it for anything sensitive, and
> never ever
> use the same password for FTP as you use for anything else...
> 'cause the
> first thing someone malicious would try would be to reuse that same
> user/pass on other sites (e.g. web mail services).
>
> All three are capable of doing Samba shares (on Windows it's just
> called
> "[windows] file&folder sharing", on *nix it's the smb:// protocol,
> and on
> Mac it is too though they mostly try to hide it). This is a relatively
> convenient method to share things when you can get the two computers
> connected. This means though that either you have to allow access
> from the
> outside and have a publicly accessible IP (or forwarded port on the
> router),
> hopefully password protected, or have the computers involved on the
> same LAN
> (e.g. a cable directly connecting them). Samba is at least moderately
> encrypted, though it is fairly easy to crack for someone with a
> modicum of
> skill.
>
> In any case, it's a viable alternate method.
>
>
> FWIW what I personally use: YouSendIt (for <100MB to normal
> people), one of
> the various servers I have access to (scp up, simple HTTP download
> down), or
> direct-cable Samba (for largish transfers between computers that
> are within
> sight of each other). YMMV and all that.
>
> - Sai
Messages in this topic (17)
________________________________________________________________________
1d. Re: OFFLIST Re: TECH: info on ftp
Posted by: "Mark J. Reed" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 2:58 am ((PDT))
1. That wasn't offlist. :)
2. No. The password isn't stored anywhere people can access it on the
site - it's not quite that bad. :)
But whenever you send information over the Internet between two
computers, software running on a third computer that is topologically
"between" the first two can look at that information, too. The fact
that such software (called a "sniffer") has to be running somewhere
between the two computers in question is not a very stringent
requirement if the two computers are far enough apart, especially if
the sniffer is installed on a lot of unsuspecting folks' computers via
a virus.
Sniffers can only capture your password in real time, while you're in
the process of logging in via FTP, but software is patient.
This is why there's ssh and scp and https:// URL's - those extra S's
stand for "secure". Which is perhaps an overstatement, but it's
definitely *more* secure. :) The data "on the wire" in such cases is
encrypted, and someone in between the two computers, even with access
to the complete session, has no way of decrypting the data.
Well, at least not as long as the cryptographic algorithm isn't
broken, quantum computing doesn't get practical, and the sniffer isn't
being run by a government or large corporation with a particular
reason to throw a lot of expensive resources at finding out David J.
Peterson's website password. But in the latter case they could
probably just pay/sue/harrass your hosting provider to get what they
need anyway. :)
On Wed, Aug 27, 2008 at 4:20 AM, David J. Peterson <[EMAIL PROTECTED]> wrote:
> Hey Sai,
>
> <<
> One thing to mind is that FTP, unless you're doing something advanced, sends
> passwords and content in the clear over the internet. This means that anyone
> between the two computers involved can read any of that data, including the
> username/password. So don't use it for anything sensitive, and never ever
> use the same password for FTP as you use for anything else... 'cause the
> first thing someone malicious would try would be to reuse that same
> user/pass on other sites (e.g. web mail services).
>>>
>
> So, I read this, and immediately thought: WHAT?! So, take
> my personal website:
>
> http://dedalvs.free.fr/
>
> I use an FTP program to update (upload stuff, etc.). (The program
> is called Transmit, by the way. It's for Mac and wonderful, but
> it costs money to register [but if you don't register it, you can
> still use it for 10 minutes at a time].) But, what exactly does this
> mean? Does this mean, essentially, that anyone that navigates
> to my site or downloads something off it can access my password
> and username? Or is this a different type of FTP transfer that
> you're talking about?
>
> -David
> *******************************************************************
> "A male love inevivi i'ala'i oku i ue pokulu'ume o heki a."
> "No eternal reward will forgive us now for wasting the dawn."
>
> -Jim Morrison
>
> http://dedalvs.free.fr/
>
--
Mark J. Reed <[EMAIL PROTECTED]>
Messages in this topic (17)
________________________________________________________________________
1e. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Lars Finsen" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 4:18 am ((PDT))
Den 27. aug. 2008 kl. 11.58 skreiv Mark J. Reed:
> 1. That wasn't offlist. :)
But it sure was of general interest.
> This is why there's ssh and scp and https:// URL's - those extra S's
> stand for "secure". Which is perhaps an overstatement, but it's
> definitely *more* secure. :) The data "on the wire" in such cases is
> encrypted, and someone in between the two computers, even with access
> to the complete session, has no way of decrypting the data.
Thanks for the info. Ok, I'd better look into this ssh and scp stuff.
Hope they're not too cumbersome. Though I do use a special password
for ftp.
LEF
Messages in this topic (17)
________________________________________________________________________
1f. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Mark J. Reed" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 5:33 am ((PDT))
Ssh is not cumbersome - there's an "sftp" that works like regular ftp
but uses ssh, so the interface looks the same. It is more flexible -
but that's a security tradeoff for the provider. If you have ssh
access you can actually execute commands on the server, in addition to
reading and writing files, which is more dangerous for them. But it
opens up opportunities to use things like rsync, which is a smart file
transfer system that only sends the parts of a file that have changed.
On 8/27/08, Lars Finsen <[EMAIL PROTECTED]> wrote:
> Den 27. aug. 2008 kl. 11.58 skreiv Mark J. Reed:
>
>> 1. That wasn't offlist. :)
>
> But it sure was of general interest.
>
>> This is why there's ssh and scp and https:// URL's - those extra S's
>> stand for "secure". Which is perhaps an overstatement, but it's
>> definitely *more* secure. :) The data "on the wire" in such cases is
>> encrypted, and someone in between the two computers, even with access
>> to the complete session, has no way of decrypting the data.
>
> Thanks for the info. Ok, I'd better look into this ssh and scp stuff.
> Hope they're not too cumbersome. Though I do use a special password
> for ftp.
>
> LEF
>
--
Sent from Gmail for mobile | mobile.google.com
Mark J. Reed <[EMAIL PROTECTED]>
Messages in this topic (17)
________________________________________________________________________
1g. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Lars Finsen" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 7:29 am ((PDT))
Den 27. aug. 2008 kl. 14.33 skrev Mark J. Reed:
> Ssh is not cumbersome - there's an "sftp" that works like regular ftp
> but uses ssh, so the interface looks the same. It is more flexible -
> but that's a security tradeoff for the provider. If you have ssh
> access you can actually execute commands on the server, in addition to
> reading and writing files, which is more dangerous for them. But it
> opens up opportunities to use things like rsync, which is a smart file
> transfer system that only sends the parts of a file that have changed.
Ok, thanks for the info. Quite a bit more to find on the internet it
seems. It's even possible to use it in Max OS X without any extra
software, although that is cumbersome. But YummyFTP, the program I
normally use for file transfer, seems to have some sftp functionality
as well.
LEF
Messages in this topic (17)
________________________________________________________________________
1h. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Keith Gaughan" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 8:18 am ((PDT))
Mark J. Reed wrote:
> Ssh is not cumbersome - there's an "sftp" that works like regular ftp
> but uses ssh, so the interface looks the same. It is more flexible -
> but that's a security tradeoff for the provider. If you have ssh
> access you can actually execute commands on the server, in addition to
> reading and writing files, which is more dangerous for them. But it
> opens up opportunities to use things like rsync, which is a smart file
> transfer system that only sends the parts of a file that have changed.
Most providers (and I work for one, so I've first-hand experience of this)
don't like providing SFTP because it requires actual real users on the
machine and requires an awful lot of infrastructure to get it to work
securely such as chroot jails and the like. Realistically, SSH and SFTP
aren't options if you've a shared hosting account, which is what the vast
majority of people would have, and anybody with the savvy to set up a
dedicated or colocated machine would also know enough to set up SSH themselves
anyway.
However, most providers will happily provide FTP over SSL/TLS aka FTPS.
FTPS is just as secure as SFTP, and because it's really just FTP (just as
HTTPS is really just HTTP over SSL/TLS), it's all that's required on the
part of the hosting provider is a valid SSL certificate assigned to the
server.
So if anybody's looking for secure file transfer between their computers
and their website, the thing to look for is if your hosting provider provides
FTPS.
K.
Messages in this topic (17)
________________________________________________________________________
1i. Re: OFFLIST Re: TECH: info on ftp
Posted by: "Sai Emrys" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 11:59 am ((PDT))
On Wed, Aug 27, 2008 at 2:58 AM, Mark J. Reed <[EMAIL PROTECTED]> wrote:
> 2. No. The password isn't stored anywhere people can access it on the
> site - it's not quite that bad. :)
>
True.
Not unless they can take advantage of a server vulnerability or the like, of
course.... and FWIW FTP servers have historically had an unusually high
number of such vulnerabilities.
(For the technically inclined: go download Metasploit, svn up, and take a
look at just how many FTP attacks there are available. Not good, eh? ;-))
There exists also sftp (SSH + FTP) and ftps (FTP + SSL). But a) if you're
using ssh, why would you want ftp rather than real scp, and b) dealing with
SSL in an actually secure manner requires either serious geekery or paying
for a real SSL certificate (man-in-the-middle attacks are trivial otherwise,
rendering the SSL useless).
But whenever you send information over the Internet between two
> computers, software running on a third computer that is topologically
> "between" the first two can look at that information, too.
Mind that, in the case that any computer along the line is on wireless,
everything it does over that wireless network is available for anyone within
physical receiving distance to read (if they can crack whatever security you
have on that wireless network).
Just as a rule of thumb: I've cracked WEP 'secured' nets in <10 min for busy
networks, 2 days for rarely used ones, and successfully connected to /
sniffed the traffic of networks ~5 miles away using good equipment on top of
a hill. I've yet to see someone crack WPA however in any practical way.
Lesson: use WPA unless you don't care if your traffic is listened in on.
> The fact that such software (called a "sniffer") has to be running
> somewhere
> between the two computers in question is not a very stringent
> requirement if the two computers are far enough apart, especially if
> the sniffer is installed on a lot of unsuspecting folks' computers via
> a virus.
FWIW as counterpoint, over-the-wire sniffing is somewhat ameliorated by the
fact that your traffic will generally take the fastest route to destination.
Which usually means that the only computers involved are: yours, anyone near
you if you're wireless, your router [it's a computer too], your ISP
frontend, backend, trunk, inter-ISP trunk, other ISP's trunk, backend,
frontend, and the website host itself (plus whatever routers, load
balancers, etc).
Relatively speaking, it's unlikely that ISP computers will be compromised
because they're generally pretty damn paranoid about it... buuuuut I know
people who work for ISPs, and have heard them talk about their machines
getting compromised because someone in the office wasn't quite as paranoid
as they ought to have been. (And as a result becoming spambots just like any
random home user... except spambots with really massive bandwidth and the
ability to monitor a few thousand home users' traffic...)
So even though it's unlikely, you should consider all traffic you send to be
suspect to being sniffed (= wiretapped / spied on) unless it's well
encrypted.
Incidentally, one thing people usually forget is that VOIP (e.g. Skype,
Vonage, etc) is usually completely unencrypted, and one can record complete
conversations in exactly the same manner. It's not difficult at all. (Good,
easy tool for Windows users: Cain & Abel. Try it out and see for yourself.)
BTW random on this: GMail defaults to only being https for login, and http
for the actual app. However, you can force it by logging in originally to
https://mail.google.com - and there's a nice plugin for Firefox, Better
Gmail 2, that does this for you automatically. (I use it.)
- Sai
Messages in this topic (17)
________________________________________________________________________
1j. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Sai Emrys" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 12:07 pm ((PDT))
On Wed, Aug 27, 2008 at 5:33 AM, Mark J. Reed <[EMAIL PROTECTED]> wrote:
> If you have ssh access you can actually execute commands on the server, in
> addition to
> reading and writing files, which is more dangerous for them.
Not necessarily - you can chroot them (i.e. restrict what they can see to a
very limited, heavily sandboxed subsystem) and set their shell to something
like /usr/bin/false or scp itself ;-)
That way, they can only do whatever you specifically allow them to do
(unless they manage to crack scp or break out of chroot, which granted has
happened but is pretty damn rare & requires serious skill).
- Sai
Messages in this topic (17)
________________________________________________________________________
1k. Re: Not OFFLIST Re: TECH: info on ftp
Posted by: "Sai Emrys" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 12:11 pm ((PDT))
On Wed, Aug 27, 2008 at 8:17 AM, Keith Gaughan <[EMAIL PROTECTED]> wrote:
> Realistically, SSH and SFTP aren't options if you've a shared hosting
> account, which is what the vast
> majority of people would have, and anybody with the savvy to set up a
> dedicated or colocated machine would also know enough to set up SSH
> themselves
> anyway.
>
FWIW: For my personal use, I have a shared hosting account w/ DreamHost. I
get full SSH etc access; I just don't have root, so I have to install stuff
within my own directory if I want to add software, and ask them to do some
things for me (like edit Apache config that can't be done in an .htaccess
file). It's reasonably cheap, though.
DH however is a provider that caters to Ruby on Rails users - for whom SSH
is not optional. For users where your site is just a bunch of HTML files
etc, not actual code, plain (S)FTP(S) is probably good enough.
Personally I'd never use it 'cause it'd feel way to restrictive, but *shrug*
- Sai
Messages in this topic (17)
________________________________________________________________________
________________________________________________________________________
2a. Re: Another Sketch: Palno
Posted by: "Veoler" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 3:14 am ((PDT))
Logan Kearsley wrote:
>> This seems to have potential for ambiguity, if one's not fastidious about
>> the positions of the commas (and maybe even if one is, with a more
>> complicated example -- I haven't convinced myself either way). Consider a
>> sentence of form
>> a who a.ACC p a.ACC p a.ACC p
>> where each a is an atom, and each p a predicate taking nom and acc
>> arguments. Does the relative clause finish after the first p, or the second?
>
> Or after the third, leaving an incomplete sentence. You can't tell
> without the commas. They are absolutely required. It would be nice to
> have a way around that, but I haven't found one yet.
Latejami (and my own conlang which is heavily influenced by Latejami) have a
parse rule that the parser won't quit the current level until it encounters
something which violates the syntax for that level. So if you have an adverb
following an embedded verb then it will always be parsed as modifying the
embedded verb and not the main one. If you want it to modify the main verb
you have to use a particle which terminates the embedded verb.
Maybe something like this could work for you language?
> Keeping up the mathematical/computational analogy, my justification
> for this structure is that a comma + relative pronoun is like a
> function call that temporarily creates a new stack frame to parse the
> next bit of the sentence in, and the closing comma is like a return
> statement that restores the original stack frame, but with the top
> element altered.
Just a question: Is this language supposed to be written-only?
--
Veoler
Messages in this topic (13)
________________________________________________________________________
2b. Re: Another Sketch: Palno
Posted by: "Veoler" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 3:17 am ((PDT))
Jim Henry wrote:
> bly-van pwÄm.
> fall-V.STATE water
>
> but some other weather verbs, like "it's hot",
> seem to want {purj} as their subject.
>
> jâln-van purj.
> hot-V.STATE environment
Why not "hot-V.STATE air"?
--
Veoler
Messages in this topic (13)
________________________________________________________________________
2c. Re: Another Sketch: Palno
Posted by: "Veoler" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 3:19 am ((PDT))
Oh, sorry, I missed you last message.
I wrote:
> Just a question: Is this language supposed to be written-only?
>
> --
> Veoler
>
Messages in this topic (13)
________________________________________________________________________
2d. Re: Another Sketch: Palno
Posted by: "Logan Kearsley" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 7:54 am ((PDT))
On Wed, Aug 27, 2008 at 6:14 AM, Veoler <[EMAIL PROTECTED]> wrote:
> Logan Kearsley wrote:
>>> This seems to have potential for ambiguity, if one's not fastidious about
>>> the positions of the commas (and maybe even if one is, with a more
>>> complicated example -- I haven't convinced myself either way). Consider a
>>> sentence of form
>>> a who a.ACC p a.ACC p a.ACC p
>>> where each a is an atom, and each p a predicate taking nom and acc
>>> arguments. Does the relative clause finish after the first p, or the
>>> second?
>>
>> Or after the third, leaving an incomplete sentence. You can't tell
>> without the commas. They are absolutely required. It would be nice to
>> have a way around that, but I haven't found one yet.
>
> Latejami (and my own conlang which is heavily influenced by Latejami) have a
*Googles Latejami*
Machine translation interlanguage?
> parse rule that the parser won't quit the current level until it encounters
> something which violates the syntax for that level. So if you have an adverb
> following an embedded verb then it will always be parsed as modifying the
> embedded verb and not the main one. If you want it to modify the main verb
> you have to use a particle which terminates the embedded verb.
> Maybe something like this could work for you language?
Hm. Yeah, I think it could, but not exactly the same, since I don't
want to have to add particles, etc. But perhaps an affix or a 1-arity
predicate that does nothing but say "this can't be an argument". And
explicit "return" statement to go along with the relative clause
function call analogy.
-l.
Messages in this topic (13)
________________________________________________________________________
2e. Re: Another Sketch: Palno
Posted by: "Jim Henry" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 8:29 am ((PDT))
On Wed, Aug 27, 2008 at 6:17 AM, Veoler <[EMAIL PROTECTED]> wrote:
> Jim Henry wrote:
>> but some other weather verbs, like "it's hot",
>> seem to want {purj} as their subject.
>>
>> jâln-van purj.
>> hot-V.STATE environment
>
> Why not "hot-V.STATE air"?
That would work too, at least in most contexts.
--
Jim Henry
http://www.pobox.com/~jimhenry/
Messages in this topic (13)
________________________________________________________________________
________________________________________________________________________
3a. Average life of a conlang
Posted by: "Jim Henry" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 8:20 am ((PDT))
Recently there's been a thread on the ZBB about the average life of a
conlang; that is, the average amount of time a conlang is under
continuing development by its creator, from initial creation to
abandonment of the conlang or death of the conlanger, not counting
auxlangs that continue to be developed by a speaker community after
the death or loss of interest of the creator.
http://www.spinnoff.com/zbb/viewtopic.php?t=28831&postdays=0&postorder=asc&start=0
I went through the thread and entered the minimum and maximum age of a
conlang mentioned for each conlanger (not each poster, as some didn't
give exact figures for themselves and some mentioned the ages of other
people's conlangs) into a tab-delimited text file, which I analyzed
with a few Awk scripts. The average minumum age of a conlang (with 32
conlangers represented) was 9.48 years; the average maximum age was
11.17 years. (20 people only gave data for their main conlang, which
in some cases is their only conlang.) The averages were skewed by a
few people with very long-lived conlangs; 11 people out of 32 had
conlangs they'd worked on for 10 years or more, and 5 had conlangs
they'd worked on for 30 years or more. 14 people had no conlang
they'd worked on for more than 3 years, but that included some people
who'd been conlanging for 3 years or less.
This data seems to fit with the pattern observed in the recent
"Promiscuity & Fidelity" thread, that conlangers tend to fall into two
types, one type working on many languages serially and another type
working on one language for years and years on end.
It would be interesting to collect the same information from members
of the CONLANG list and get a broader base of data on the question.
Jörg and I have already posted about our conlangs on the ZBB thread,
but I'll copy my posts here as ZBB threads disappear into the bit
bucket after a few months (or even weeks?) and CONLANG threads
get archived indefinitely:
===
halyihev wrote:
> I started Alurhsa in January of 1977. So that makes it 31
> and a half.
> I think Sally Caves' Teonaht was started sometime in the
> early 1960's, though. That's the oldest one I know.
Dr. Peter Tarlow, Paul Burgess, and Bill Price also have languages
they've been working on since they were children or teenagers, roughly
as old as Teonaht plus or minus a few years. Tarlow's La Petro was
fifty years old when he made his first introductory post to the
CONLANG list last September.
Tolkien worked on Quenya, Noldorin and Sindarin pretty much from
about 1917 to his death in 1973, right? -- fifty-six years, pretty
respectable.
My gjâ-zym-byn is young compared to your Alurhsa and the other
conlangs I mentioned, but probably older than the average
continually-developed conlang; it turned 10 in early March.
Three of my other conlangs I've worked on for 1-3 years; another dozen
or so been sketchy projects that I've worked on for less than a year,
in some cases less than a week. One of those I worked on for a couple
of years I seriously intend to get back to working on sometime;
another is still in progress, and the oldest has been abandoned pretty
much since I started seriously working on gzb in early 1998.
===
Volapük too is still being updated by its tiny remnant speaker
community, with new words for new concepts; not as vigorously as
Esperanto, but it's not dead, just pining. We need to make a
distinction between conlangs that are developed/used by one person,
and those that are used (and necessarily continually developed, if
they're to be used very much) by a speaker community which might
survive the creator. Lingua Ignota has been accumulating dust ever
since Hildegard died; no one else has really used it. (We're not sure
exactly how much Hildegard herself used it, or how.) Ditto with most
of the sketchy philosphical languages and auxlangs of the 17th-19th
centuries. Their "age" should be reckoned, in the context of this
thread, as the period of time their creators continually worked on
them (unknown in most cases), not the period of time from their first
publication to the present. Whereas in the case of auxlangs with a
speaker community, their age should maybe be reckoned from the date
the *second* person (after the creator) learned the language fluently
- for Esperanto, I think that would be Antoni Grabowski in late 1887.
===
Tsiasuk-Pron wrote:
> Itlani is eleven years old. For me, the point when Itlani reached a
> stage of real stability was the true beginning of the language. For
> me, languages are tools of self-expression. If I kept on building
> them to the point where they were able to be used and then
> abandoning them it would not be as much fun.
Do you mean, Itlani reached stability eleven years ago and you started
it at some earlier point, or that you started creating it eleven years
ago and it reached stability at some later point?
gjâ-zym-byn reached a usable level of stability in about 2000 or 2001,
after two or three years of development. There have been many changes
since then, but relatively little that's not backward compatible with
the existing corpus. (Except for archaicizing old words; there are
over 50 words in the dictionary marked as archaic, and some older ones
that have no dictionary entry.) It was roughly around 2002-2004 when I
became fluent enough in the written form of the language that I could
write about as fast in gzb as in English, on the subjects I normally
talk about in my journal. (I'm still not truly fluent in the spoken
language; though I think in it and talk to myself in it fairly often,
it's with far more hesitations and disfluencies than in English or
Esperanto.)
And yet I don't regret the time I spent working on short-lived sketchy
conlangs, either; most of them have a respectable corpus relative to
their short lifespan, they explored interesting linguistic ideas, and
I think the availability of those shorter-lived projects has diverted
my tinkering tendencies away from gzb and made it stable enough to
learn.
--
Jim Henry
http://www.pobox.com/~jimhenry/conlang/fluency-survey.html
Conlang fluency survey -- there's still time to participate before
I analyze the results and write the article
Messages in this topic (2)
________________________________________________________________________
3b. Re: Average life of a conlang
Posted by: "Herman Miller" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 5:48 pm ((PDT))
Jim Henry wrote:
> Recently there's been a thread on the ZBB about the average life of a
> conlang; that is, the average amount of time a conlang is under
> continuing development by its creator, from initial creation to
> abandonment of the conlang or death of the conlanger, not counting
> auxlangs that continue to be developed by a speaker community after
> the death or loss of interest of the creator.
I don't have accurate figures for most of my languages, but the average
would be misleading as it would include a vast number of sketchy langs
with a lifetime of less than a year. I guess you could call them
mouselangs. My first conlang, Olaetian, was under development for around
15 years, from the late 1970's to the mid 1990's. I've been developing
Tirelat since 1999, but since I started developing Minza in late 2004
until recently, Tirelat has been on hold for most of that time. It's
always possible that I could revive older languages from the mid-1980's
which have been dormant all these years (but not likely).
Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
4a. Playing with dialect variations
Posted by: "Logan Kearsley" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 8:48 am ((PDT))
Thinking about English dialects and lazy speech, I came up with two
modifications to English, and I wonder if the relevant features occur
in any natural language (or, if not, in any conlangs).
First- single particles combining tense, aspect, and implied subject.
Illustrated with the sentence "I'm going to jump off the roof."
-> "I'm gonna jump off the roof" [The infinitive particle has fused
with the tense & aspect marking gerund.]
-> "I'ma jump off the roof" [Not sure if that's a contraction
eliminating 'gonn' or an addition of the archaic progressive particle
"a-" as in "a-building" and then dropping "gonna".]
-> "Ima jump off the roof" or just "Ma jump off the roof", since the
"m" encodes the same subject information as the "I".
The original verb "am" has been completely contracted away along with
the infinitive particle, so we can re-analyze "jump" as a finite verb
modified by a particle "ima" which encodes tense, aspect, and the
first person subject.
You can go through a similar process with other subjects, to get
particles like "esa/shesa" (he/she is going to), "yura" (you are going
to), "theyra", etc. And when the implied subjects encoded in those
particles aren't sufficient, then you'll need agreement between the
particle and the subject. Sort of like verb conjugation, except that
it's a separate particle altering, not the verb itself. The agreement
is already set up in dialects that use constructions like "Bobby, he's
going to jump off the roof", where "Bobby" refers to the subject, not
the addressee (in which case, a vocative particle like "yo", or "hey"
is required to disambiguate the addressee). That conveniently
transforms into "Bobby esa jump off the roof."
Second- verbs as a closed class.
Take the afro-american dialect example "I done gone" for "I left" or
"I am gone", or "I done went to the store" for "I went to the store,
but now I'm back again". "Done" is actually being used as an aspect
particle, rather than as a verb, but what if I re-analyze "done" as
the verb and "gone" or "went" as its object?
You get a situation where most everything that is expressed as a verb
in English becomes a noun for the action of that verb, and verbs
become a fairly small closed class (to be, to do, to become, probably
a few others).
"I go" becomes "I do go" / "I do going"
"I work" becomes "I do work"- same as in normal English, since "work"
is both a noun and a verb.
-l.
Messages in this topic (2)
________________________________________________________________________
4b. Re: Playing with dialect variations
Posted by: "Philip Newton" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 9:38 am ((PDT))
On Wed, Aug 27, 2008 at 17:48, Logan Kearsley <[EMAIL PROTECTED]> wrote:
> Second- verbs as a closed class.
>
> Take the afro-american dialect example "I done gone" for "I left" or
> "I am gone", or "I done went to the store" for "I went to the store,
> but now I'm back again". "Done" is actually being used as an aspect
> particle, rather than as a verb, but what if I re-analyze "done" as
> the verb and "gone" or "went" as its object?
> You get a situation where most everything that is expressed as a verb
> in English becomes a noun for the action of that verb, and verbs
> become a fairly small closed class (to be, to do, to become, probably
> a few others).
>
> "I go" becomes "I do go" / "I do going"
> "I work" becomes "I do work"- same as in normal English, since "work"
> is both a noun and a verb.
Looks a lot like Basque to me.
Cheers,
--
Philip Newton <[EMAIL PROTECTED]>
Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
5a. Re: Language Sketch: Gogido
Posted by: "Philip Newton" [EMAIL PROTECTED]
Date: Wed Aug 27, 2008 10:05 am ((PDT))
On Tue, Aug 26, 2008 at 01:20, Andreas Johansson <[EMAIL PROTECTED]> wrote:
> Quoting "Mark J. Reed" <[EMAIL PROTECTED]>:
>
>> On Mon, Aug 25, 2008 at 6:03 PM, Logan Kearsley <[EMAIL PROTECTED]>
>> wrote:
>> > It implies that my default pronunciation in most cases is voiced,
>> > although there's free variation between the voiced and unvoiced
>> > allophones.
>>
>> In the case of Okaikiar, I'm not consistent. It has [t] and [d] in
>> allophonic variation, likewise [s] and [z], but there's no [g], only
>> [k]. I suspect this is massively unrealistic.
>
> There are languages, most famously perhaps varieties of Arabic, that have /t d
> k/ but no /g/ or [g]*. Given this, having a voiced allophone of /t/ but not of
> /k/ doesn't seem surprising in a language that doesn't distinguish phonemic
> voice.
>
> * Having /t d g/ but no /k/ is apparently less common - the explanation is
> supposed to be that modal voicing is relatively hard to sustain for back stops
> (pressure builds up faster due to less space between the glottis and the
> closure
> of the vocal track) so that if there's only back stop it tends to "default" to
> voicelessness.
And at the other end of the buccal tract, apparently if there is only
one bilabial stop, it tends to be a voiced one (Arabic is another
example here, with /b/ but no /p/).
Cheers,
--
Philip Newton <[EMAIL PROTECTED]>
Messages in this topic (14)
------------------------------------------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/conlang/
<*> Your email settings:
Digest Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/conlang/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:[EMAIL PROTECTED]
mailto:[EMAIL PROTECTED]
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
------------------------------------------------------------------------
