Re: [Haifux] [W2L] Call for lecturer + Linux guru
Tzafrir Cohen wrote: I'm not sure I agree with you regarding version control systems. Specifically distributed version control systems make the common case of a repository for the project simple. Unlike Subversion, you don't need to set up a separate server. You do not need to set up a separate server in subversion. svnadmin create ~/svn ; svn co file:///home/sun/svn is enough. And it saves you a whole lot of time in saving ex1.c_1 , ex1.c_2, ex.c.orig and such. Has anyone compared git's performance with file:// based svn? I'm not sure that claim holds. I think that demonstarting simple linear workflow (no branches, no remote repositories) with git, bzr or hg could be handy. Guy has already responded with some shooting yourself in the foot, but I'll elaborate. Git makes it exceedingly easy to reach situations in which information is either actually lost, or is not technically lost, but is unreachable without searching in internal log files. The most problematic one to chance upon by accident is off branch commits (where you revert to an old version, make changes and commit them by mistake). I read somewhere an explanation of how to recover from this situation. It started by explaining how lucky he was to have spotted that that was what he did before he checked out another branch (if you do, the only way to find your commits is the afore mentioned internal log files). He then goes on to list the steps required to recover the file, and ends with viola! it's like magic, with a footnote stating that like magic means it is a list of arbitrary and hard to remember incantations in which getting the slightest thing wrong means a catastrophe. At least the other case, where data is actually lost, is unlikely to get at if you only work with one repository[1]. In short, I'm with Guy on this one. Git is an extremely powerful tool, but it is only useful once you master the right way to tread and where. In other words - not a good choice for a novice[2]. Shachar 1 The rebase operation actually changes the history. After a branch rebase, the tree is reworked to look as if the branch has always branched off the new rebase point. This means that the precise tree layout before the rebase operation is no longer reachable. It is no longer possible to check out the version that compiled fine before the rebase. I don't remember the precise details at the moment to say whether every rebase is like that, or whether it's only sometimes. I will say that, in my book, a version control that allows an operation that is not undoable is not a version control. As I write this, it occures to me that this may actually be the precise same problem as the one I pointed above, and the previous end point for the branch is still in the repository, it just doesn't have a name, and if you know the commit's name, you can still check it out. That does not change my above statement - it is a no-no for a version control system. 2 In my opinion, also not a good choice for anyone who relies on centralized backups (like a commercial company), not a good choice for anyone who does not perform multi-commiters multi-versions development. In other words, there are very few cases, aside from the Linux kernel itself, where it IS a good choice. Also, I often find where even where it is a good choice (such as Wine, where an off trunk personal development branch might be long and convulted), it is not used properly (the development model does not support a git pull and multi commiters - only patch sending). In the android case, it is the best choice available (multiple repositories, independently developed in parallel, with partial code sharing between them), but has significant drawbacks (no support for multiple repository checkouts, no support for partial tree checkouts) which are worked around with a wrapper tool. -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
Vadim Eisenberg wrote: Eclipse doesn't belong to the right tools, in my opinion. Why Eclipse doesn't belong to the right tools ? My naïve understanding is that Eclipse is Emacs of the 21-st century – it is open source, customizable etc., similar to Emacs; in addition to being graphical. Thank you! I was wondering why I hated Eclipse so much, and you have put your finger on it. It's exactly like a 21-st century Emacs. Shachar p.s. I'm not so sure it is a bad idea to teach Eclipse. I think it's worth mentioning that eclipse exists, but to just point them in the right direction, and then teach the basics. -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
[Haifux] OT: yes, I do want duplicates
Slightly off-topic, but: I noticed that messages from the recent discussion were only coming into my INBOX and not into my haifux folder. Fix: logging into the mailman web interface and setting Avoid duplicate copies of messages? to No. -- Tzafrir Cohen | tzaf...@jabber.org | VIM is http://tzafrir.org.il || a Mutt's tzaf...@cohens.org.il || best ICQ# 16849754 || friend ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
guy keren wrote: you can mention memory leaks if you want - but students don't care about them so much, because it doesn't break their programs. Starting from the Winter 2008-2009 semester, the memory leaks are checked in Matam (Introduction to Systems Programming) course and 1 point is reduced for each leak. The check is done automatically, using, guess what, valgrind. So the students in Matam actually care very much about the memory leaks. Vadim -Original Message- From: haifux-boun...@haifux.org [mailto:haifux-boun...@haifux.org] On Behalf Of guy keren Sent: Friday, October 16, 2009 9:47 AM To: Eli Billauer Cc: Haifa Linux Club Subject: Re: [Haifux] [W2L] Call for lecturer + Linux guru Eli Billauer wrote: guy keren wrote: what - no valgrind? I stand corrected. A quick demonstration of valgrind (show how it detects memory leaks and access to unallocated/uninitialized memory) is in place. It's definitely something handy for a student, and it's so simple to use. Eli i think the best demo for valgrind would be: 1. this step should be done at home: write a program that has a non-obvious problem with corrupting its memory (make sure that when you run it, it actually crashes) 2. the next steps will be done during the demonstration: show it to the people (the source, how the program crashes, and how you fail to find the problem even when the crash is done inside gdb) 3. show them how you find the bug within seconds using valgrind. you can mention memory leaks if you want - but students don't care about them so much, because it doesn't break their programs. they do care about memory corruption if it indeed causes their program to crash. --guy ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
it seems that on t2 (now called stud) you can use not only file:/// but also svn+ssh://, with only a student account. in that case i agree that subversion is much better than git for this purpose. as for the question of why do you need version control for something as simple as MATAM, it really makes a huge difference from my experience. it saves alot of the ok i did this now you do this and the crap we messed up let's hope there's a backup of this in my inbox and lets you focus on the work. while some students may be skeptical, this IMHO will become a killer app and a must for many students and may give us an upper hand. On Fri, Oct 16, 2009 at 10:05, Vadim Eisenberg vadim.eisenb...@gmail.comwrote: guy keren wrote: you can mention memory leaks if you want - but students don't care about them so much, because it doesn't break their programs. Starting from the Winter 2008-2009 semester, the memory leaks are checked in Matam (Introduction to Systems Programming) course and 1 point is reduced for each leak. The check is done automatically, using, guess what, valgrind. So the students in Matam actually care very much about the memory leaks. Vadim -Original Message- From: haifux-boun...@haifux.org [mailto:haifux-boun...@haifux.org] On Behalf Of guy keren Sent: Friday, October 16, 2009 9:47 AM To: Eli Billauer Cc: Haifa Linux Club Subject: Re: [Haifux] [W2L] Call for lecturer + Linux guru Eli Billauer wrote: guy keren wrote: what - no valgrind? I stand corrected. A quick demonstration of valgrind (show how it detects memory leaks and access to unallocated/uninitialized memory) is in place. It's definitely something handy for a student, and it's so simple to use. Eli i think the best demo for valgrind would be: 1. this step should be done at home: write a program that has a non-obvious problem with corrupting its memory (make sure that when you run it, it actually crashes) 2. the next steps will be done during the demonstration: show it to the people (the source, how the program crashes, and how you fail to find the problem even when the crash is done inside gdb) 3. show them how you find the bug within seconds using valgrind. you can mention memory leaks if you want - but students don't care about them so much, because it doesn't break their programs. they do care about memory corruption if it indeed causes their program to crash. --guy ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux -- Mike Ditka http://www.brainyquote.com/quotes/authors/m/mike_ditka.html - If God had wanted man to play soccer, he wouldn't have given us arms. ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
I tend to agree that git is somewhat complex for the novice. It takes awhile before one feels comfortable working with git. SCMs in general, are not considered core learning material, so most student will prefer to avoid wasting their time to earn the required expertise. Having said that, for the experienced git users, there is not reason to lose their work, even in the case of some mistake. For example, most problems can be recovered using: http://www.kernel.org/pub/software/scm/git/docs/user-manual.html#recovering-lost-changes Another, less powerful tool, but good enough for immediate disaster recovery is the ORIG_HEAD reference. --doron On Thu, Oct 15, 2009 at 11:38 PM, guy keren c...@actcom.co.il wrote: the problem with git, is that it's very easy to shoot yourself in the foot. giving it to students, who might accidentally reset their repository into losing their code, is not a very good idea, if you don't have time to give a proper explanation plus warnings. --guy Tzafrir Cohen wrote: On Thu, Oct 15, 2009 at 09:13:58PM +0200, Eli Billauer wrote: OK, I think this is a good time to express my view regarding the Development tools lecture. It's purpose, as I see it, is to give the students a nice start with the right tools for developing code, as needed for their exercises. If their experience is good, they'll stay. If not, they'll soon use the alternatives. If you want to give a lecture about any other subject, as a Stay-in-Linux or mainstream lecture, by all means come forward. But let's try to get some focus on the initial lecture. Correct me if I'm wrong, but a student is not likely to go beyond a project which runs on a single platform, having a few source files, and with no more than two or three persons involved. Hence autotools are irrelevant, and so are version control systems. Tarballing all sources, and sending to your partner with comments, is as much version control as you need in these situations. I'm not sure I agree with you regarding version control systems. Specifically distributed version control systems make the common case of a repository for the project simple. Unlike Subversion, you don't need to set up a separate server. And it saves you a whole lot of time in saving ex1.c_1 , ex1.c_2, ex.c.orig and such. I think that demonstarting simple linear workflow (no branches, no remote repositories) with git, bzr or hg could be handy. ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
I tend to disagree about git being too complex. I currently have three students (in a military setting), which have never used any form of version control before, and have been taught basic usage of git - init, add, commit, log, diff, remote, and pull. I've received no complaints as of yet. As long as you're not doing rebases or working with multiple branches (which are much more complicated to do in SVN, and useless in the situation at any rate), the data loss problems mentioned above don't exist. Git gives the added bonus of being able to work offline, which is indispensable for a student on a laptop. -- Man is the only animal that laughs and weeps, for he is the only animal that is struck with the difference between what things are and what they ought to be. - William Hazlitt Ohad Lutzky ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
On Fri, Oct 16, 2009 at 10:05:14AM +0200, Vadim Eisenberg wrote: guy keren wrote: you can mention memory leaks if you want - but students don't care about them so much, because it doesn't break their programs. Starting from the Winter 2008-2009 semester, the memory leaks are checked in Matam (Introduction to Systems Programming) course and 1 point is reduced for each leak. The check is done automatically, using, guess what, valgrind. So the students in Matam actually care very much about the memory leaks. ... which are detectable by valgrind :-) -- Tzafrir Cohen | tzaf...@jabber.org | VIM is http://tzafrir.org.il || a Mutt's tzaf...@cohens.org.il || best ICQ# 16849754 || friend ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
On Fri, Oct 16, 2009 at 01:35:09PM +0200, boazg wrote: it seems that on t2 (now called stud) you can use not only file:/// but also svn+ssh://, with only a student account. in that case i agree that subversion is much better than git for this purpose. git works just as well with git-over-ssh repositories. Alternatively you can also use the http method to provide a public repository off t2 from your homepage :-) (Just make sure to 'chown +x .git/hooks/post-update' At least here - Debian Lenny - this makes a repository run the required git-update-server-info) -- Tzafrir Cohen | tzaf...@jabber.org | VIM is http://tzafrir.org.il || a Mutt's tzaf...@cohens.org.il || best ICQ# 16849754 || friend ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
How about a regular haifux slot, dedicated to source control applications wars? On Fri, Oct 16, 2009 at 5:30 PM, Ohad Lutzky o...@lutzky.net wrote: I tend to disagree about git being too complex. I currently have three students (in a military setting), which have never used any form of version control before, and have been taught basic usage of git - init, add, commit, log, diff, remote, and pull. I've received no complaints as of yet. As long as you're not doing rebases or working with multiple branches (which are much more complicated to do in SVN, and useless in the situation at any rate), the data loss problems mentioned above don't exist. Git gives the added bonus of being able to work offline, which is indispensable for a student on a laptop. -- Man is the only animal that laughs and weeps, for he is the only animal that is struck with the difference between what things are and what they ought to be. - William Hazlitt Ohad Lutzky ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
Doron Tal wrote: Having said that, for the experienced git users, there is not reason to lose their work, even in the case of some mistake. For example, most problems can be recovered using: http://www.kernel.org/pub/software/scm/git/docs/user-manual.html#recovering-lost-changes I have to say I am even more worried than I was. It says there that any standard maintenance will destroy that data permanently. That is, actually, worse than I thought. To me, that is a must never happen situation. Shachar -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
Ohad Lutzky wrote: As long as you're not doing rebases or working with multiple branches (which are much more complicated to do in SVN, and useless in the situation at any rate), the data loss problems mentioned above don't exist. Not true. The problem can happen if you just check out a commit which is not at the head of a branch (say, because you were doing regression testing), and then perform a commit. No branching required. Git gives the added bonus of being able to work offline, which is indispensable for a student on a laptop. True, but it steps up the complexity considerably. Well, you step it up to begin with when you say that publishing your work requires two stages (three if you count the add) - commit and then push. This means that for several people working on one repository, you would usually try to simplify things by telling them to git add + commit + push. Then, when you try take the laptop away, you find that you split an operation that used to be atomic, and the complexity comes back. True, working with a centralized repository means that this is downright impossible, but I still maintain that working offline is no novice task. Shachar -- Shachar Shemesh Lingnu Open Source Consulting Ltd. http://www.lingnu.com ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
Re: [Haifux] [W2L] Call for lecturer + Linux guru
I specifically didn't teach them checkout, for this exact reason... Yes, warnings about these things are in order when you're using git. (Specifically we have always mind your current branch and rebasing is a destructive operation, but also you can always fix these things if you notice early enough, early enough being a rather loose from my experience). On the specific matter of non-branch commits, git does some work to actively warn you when you do so, but I think this could be improved. I still believe that with warning on those two issues, git is simple enough to use, and that the ability to work offline is well worth it. On Fri, Oct 16, 2009 at 9:53 PM, Shachar Shemesh shac...@shemesh.bizwrote: Ohad Lutzky wrote: As long as you're not doing rebases or working with multiple branches (which are much more complicated to do in SVN, and useless in the situation at any rate), the data loss problems mentioned above don't exist. Not true. The problem can happen if you just check out a commit which is not at the head of a branch (say, because you were doing regression testing), and then perform a commit. No branching required. Git gives the added bonus of being able to work offline, which is indispensable for a student on a laptop. True, but it steps up the complexity considerably. Well, you step it up to begin with when you say that publishing your work requires two stages (three if you count the add) - commit and then push. This means that for several people working on one repository, you would usually try to simplify things by telling them to git add + commit + push. Then, when you try take the laptop away, you find that you split an operation that used to be atomic, and the complexity comes back. True, working with a centralized repository means that this is downright impossible, but I still maintain that working offline is no novice task. Shachar -- Shachar Shemesh Lingnu Open Source Consulting Ltd.http://www.lingnu.com -- Man is the only animal that laughs and weeps, for he is the only animal that is struck with the difference between what things are and what they ought to be. - William Hazlitt Ohad Lutzky ___ Haifux mailing list Haifux@haifux.org http://hamakor.org.il/cgi-bin/mailman/listinfo/haifux
