- implementations-list - lists.ourproject.org

[implementations-list] About bugs
by Alessandro Piras 23 Jun '09

23 Jun '09

Dear users and contributors, I had the feeling these days that undos don't work properly. I was almost sure it sucked big time on certain occasions, but in this last hour I was trying to reproduce the bad behavior and everything worked properly. If you find undo behaving in a non-natural way, please report it here on the mailing list, with proper instructions to reproduce the bad behavior. Thank you, Alessandro.

1 0

[implementations-list] Maciej's from-scratch Vim emulation project (was: Transferring maintainership of Vimpulse)
by Jason Spiro 22 Jun '09

22 Jun '09

On Thu, Jan 29, 2009 at 1:50 AM, Jason Spiro<jasonspiro3(a)gmail.com> wrote: > On Wed, Jan 28, 2009 at 6:14 PM, Stephen Bach <sjbach(a)sjbach.com> wrote: > ... >> I agree there are problems with viper, but I'm not positive replacing it >> wholesale is the best approach. In my experience, the more ambitious >> the project, the less likely it is to reach any milestone of usefulness. >> Viper is 13000+ lines of seasoned code, and it mostly works in its goal >> of pure-vi emulation. > > I agree. It would be very hard to rewrite Viper from scratch. As > popular blogger Joel Spolsky wrote: > > "The idea that new code is better than old is patently absurd. > Old code has been used. It has been tested. Lots of bugs have been > found, and they've been fixed. ... [Architectural] problems can be > solved, one at a time, by carefully moving code, refactoring, changing > interfaces. Even fairly major architectural changes can be done > without throwing away the code. On the Juno project we spent several > months rearchitecting at one point: just moving things around, > cleaning them up, creating base classes that made sense, and creating > sharp interfaces between the modules. But we did it carefully, with > our existing code base, and we didn't introduce new bugs or throw away > working code."[1] > > The books "Working with Legacy Code" by Michael Feathers and > "Refactoring" by Martin Fowler, and the SourceMaking website[2], talk > about how to refactor. The free PDF ebook "Object-Oriented > Reengineering Patterns"[3] also talks about dealing with old code, > though I think "Working with Legacy Code" is better. > > ... >> - You seem to know what you're doing, so good luck. I'll follow your >> work, and when vimp / vimpersonator is ready, I'll try it. :-) > > Yes; if you go ahead and write it, then when it's ready, I too will > try it. Good luck. > > -Jason > > ^ [1]. http://www.joelonsoftware.com/articles/fog0000000069.html > > ^ [2]. http://sourcemaking.com/refactoring > > ^ [3]. http://www.iam.unibe.ch/~scg/OORP/ Hi Maciej, I was discussing other vi clones for Emacs with Alessandro over Gtalk today. I mentioned John J. Foerch's http://jjfoerch.com/modal-mode/ and your project. How many lines of code is Vimpersonator / Vimp / whatever it's called? How well does it work so far? Could you please post a copy in some public SVN/Git-hosting site online like www.assembla.com so we could see it? -Jason

1 0

[implementations-list] Moving to Assembla (was: Which version control system should we use: darcs / git / svn / hg / other)
by Jason Spiro 22 Jun '09

22 Jun '09

On Mon, Jun 22, 2009 at 5:29 PM, Alessandro Piras<laynor(a)gmail.com> wrote: > would be ok for me :) > Is assembla any better than sourceforge? I see you're suggestig to move the > project to assembla, despite the visibility sourceforge would give. what > makes assembla a winner over sf? Trac makes it a winner. Trac is far better than the Sourceforge bug tracker. Did you yet respond to the Assembla invitation email I sent you? :)

2 1

[implementations-list] Which version control system should we use: darcs / git / svn / hg / other (was: Vimpulse update)
by Jason Spiro 22 Jun '09

22 Jun '09

On Mon, Jun 22, 2009 at 1:58 AM, Alessandro Piras <laynor(a)gmail.com> wrote: > yeah, I was thinking of getting some kind of repository for the code, > as using only emacs-wiki is difficult also for me. Is git easy to use? > I was thinking about darcs, that shouldn't bee too complicated. ... I think it'd be best to use something that: 1. we can find free online hosting for. 2. works with Assembla, since Vimpulse's bug tracking is on Assembla already. So this means svn, git, or Mercurial. Darcs is not supported. 3. a lot of people already know the basics of how to use. 4. is easy to learn, so that if new maintainers or co-maintainers step up, they can continue easily. Based on what I have read, svn and Mercurial are easy to learn, but Git is harder. But I am not a maintainer. So it is 100% your choice what to use. What do you think you will use? P.S. I already know how to use svn and git, but I learned them a long time ago and don't remember how hard they are to learn.

2 3

[implementations-list] Vimpulse moved on sourceforge
by Alessandro Piras 22 Jun '09

22 Jun '09

Dear users and contributors, Vimpulse has moved on sourceforge: https://sourceforge.net/projects/vimpulse/ I'm new to this stuff, so the project management will be clumsy for a while. The project is now split up in several files, that get concatenated thru a simple Makefile (just 1 cat command). I don't like this approach very much, and I'd rather split Vimpulse in more files, but doing things this way makes it easy to download and use it, and release the stuff also on emacswiki. what do you think? At present time, I just split the comments in various files, like TODO, Acknowledgements, README, INSTALL and one big code file, named vimpulse-code.el. I'm thinking about splitting the code too, as I don't like to work with huge files. What do you think? cheers, Alessandro

1 0

Re: [implementations-list] Vimpulse update
by Marius Andersen 22 Jun '09

22 Jun '09

> Fra: Alessandro Piras <laynor(a)gmail.com> > Emne: Re: [implementations-list] Vimpulse update > Til: "Stephen Bach" <sjbach(a)sjbach.com>, "Vi/Vim implementations list" <implementations-list(a)lists.ourproject.org> > Dato: Mandag 22. juni 2009 03.58 > > By the way, new update already, now there's partial text > object support. Try some di(, da[, diw, das dip commands and tell > me what you think. Alessandro, I must say I am impressed of the pace with which you implements things! I had a go at text objects myself some months ago, but never got any further than redefining the gigantic `viper-prefix-arg-com' function. Out of curiosity, what are the technical details behind the new support? _________________________________________________________ Alt i ett. Få Yahoo! Mail med adressekartotek, kalender og notisblokk. http://no.mail.yahoo.com

2 1

[implementations-list] Alessandro, I think we should get people to agree to assign copyright. Do you?
by Jason Spiro 22 Jun '09

22 Jun '09

Hi Alessandro, The FSF requires that all contributors sign a form agreeing to assign their copyright to the FSF before the code can get into Emacs. The FSF copyright clerk, Mr. Robertson, said that any code contribution 15 lines or more might be copyright-significant. In the thread I just forwarded to the list, you can see that he said that. I think we should get contributors to agree that they will send a copyright assignment form to the FSF when, hopefully, Vimpulse will get into Emacs. Otherwise, if you ask a contributor to assign copyright and they don't agree, it could be a big headache. I assume you would have to delete the code they contributed plus all code that requires the deleted code. I propose this. 1. I edit vimpulse.el saying how to submit patches: a) attach a patch to Trac b) mention that you are willing to assign copyright to the GNU Project if we ask you to later 2. If someone submits a patch 15 lines or more and forgot to do 1b), you ask them, and if they don't agree, you ignore the patch. You are maintainer, so this is only a suggestion to you. What do you think? By the way, I have asked at http://thread.gmane.org/gmane.emacs.devel/108538 if Vimpulse can be shipped with Emacs. We will see what they say. -Jason

3 4

Re: [implementations-list] Vimpulse update
by Alessandro Piras 22 Jun '09

22 Jun '09

yeah, I was thinking of getting some kind of repository for the code, as using only emacs-wiki is difficult also for me. Is git easy to use? I was thinking about darcs, that shouldn't bee too complicated. By the way, new update already, now there's partial text object support. Try some di(, da[, diw, das dip commands and tell me what you think. Alessandro 2009/6/21 Stephen Bach <sjbach(a)sjbach.com>: > On Thu, Jun 18, 2009 at 01:49:14PM -0400, Stephen Bach wrote: >> Hi Alessandro, >> >> On Wed, Jun 17, 2009 at 11:50:54PM +0200, Alessandro Piras wrote: >> > I'm going to upload the stuff on emacs wiki when I discover how to. >> >> Maybe instead (or in addition) you could consider opening a public >> repository for Vimpulse, perhaps on Github? It would be good to have a >> released reference version of Vimpulse on EmacsWiki, but it's difficult >> to track changes. A public repository could encourage contribution. > > Alessandro, have you given any thought to this suggestion? > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > > iEYEARECAAYFAko+frMACgkQ6gBF1Z6ARyy2wQCeOQFHr2aoljscKiA83eBW8R7n > uXAAn1x904WLWqzbN3/dPfotf8dWmyQx > =Sdng > -----END PGP SIGNATURE----- > >

1 0

[implementations-list] Alessandro, I'm done forwarding Vimpulse patches. Also want to take over viper-in-more-modes maintenance?
by Jason Spiro 21 Jun '09

21 Jun '09

Hey Alessandro, I've finished forwarding all still-applicable-looking Vimpulse patches to http://my-trac.assembla.com/vimpulse/report/1 . You can read them all there. I didn't do anything about any viper-in-more-modes patches I found in my inbox. Do you want to take over viper-in-more-modes maintenance too? If so, I'll create a Trac for viper-in-more-modes. Cheers, -Jason -- Jason Spiro: software/web developer, packager, trainer, IT consultant. I support Linux, UNIX, Windows, and more. Contact me to discuss your needs. +1 (416) 992-3445 / www.jspiro.com

1 0

[implementations-list] Fwd: [gnu.org #393190] Dear FSF clerk, here are all my questions about copyright assigment
by Jason Spiro 21 Jun '09

21 Jun '09

Forwarded conversation Subject: Dear FSF clerk, here are all my questions about copyright assigment ------------------------ From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Mon, Dec 1, 2008 at 1:40 AM To: assignments(a)gnu.org Cc: "Vi/Vim implementations: developer discussion and user tech support" <implementations-list(a)lists.ourproject.org> Dear FSF copyright assignment clerk, I am the maintainer of two Emacs add-ons: Vimpulse[1] and viper-in-more-modes[2]. Vimpulse is still too buggy to ship with Emacs, and I don't use or care much about viper-in-more-modes so have never even evaluated whether they can ship with Emacs. But I would like to start getting copyright assignments for both add-ons now so that they can ship with Emacs eventually. I care so little about viper-in-more-modes that I will not talk about it in the rest of this email. The rest of this email talks about Vimpulse only. I am not a main contributor. There are two main contributors: Alessandro Piras (in Italy) and Brad Beveridge (I think in Canada). They said they are both willing to sign copyright assignment papers. I have contributed maybe 5 lines of code and dozens of lines of documentation. I have already disclaimed all copyright to my contributions and placed my contributions in the public domain, but I will gladly sign papers too if you like. Question 1: Is it good that I have placed my contributions in the public domain? (I live in Canada.) There have been some other contributors. I have no idea exactly how many; I would have to look at each of the 80 old email threads in my inbox relating to Vimpulse. Maybe I will move them all into the Vimpulse bug tracker at ourproject.org, which is currently empty, for ease of future tracking. The following are estimates based on memories from last year: Some people have contributed one- or two-line fixes. Perhaps half a dozen (or maybe less) people have contributed larger amounts of code: e.g. ten to twenty lines. Question 2: What is the minimum number of lines that requires copyright assignment? (I tried doing a Google search for 'fsf copyright assignment' and found the old document [3] from 'prep' which refers to various files that let me check whether papers have been received and talk more about the various options (assign vs. disclaim) and such. But I couldn't find the files it refers to. It would be wonderful if you could put those files up on the web. Q3: What if someone contributes code which does nothing but define keybindings (keyboard shortcuts)? Q4: Should I include lines of documentation / comments in my count of lines of code? How about blank lines? Q5: There are several unreviewed patches in my inbox. Do you recommend I leave them unreviewed until I have gotten copyright assignment for them? Q6. How about if I tell the pending patch contributors, "Please contact assignments @ gnu.org so you can assign your contributions to the FSF (this is best) or disclaim copyright to your contributions and put them in the public domain (if you are unwilling to assign copyright). After that, please test your patch, and if it tests OK, please merge it into Vimpulse on the EmacsWiki." Please Reply to All. Thanks in advance, --Jason Spiro ^ [1]. http://www.emacswiki.org/elisp/vimpulse.el ^ [2]. http://www.emacswiki.org/emacs/viper-in-more-modes.el ^ [3]. http://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html ---------- From: Donald R Robertson III via RT <copyright-clerk(a)fsf.org> Date: Mon, Dec 1, 2008 at 6:16 PM To: jasonspiro3(a)gmail.com Hello, It is great to hear that you are interested in contributing to Emacs. There's actually a bit of process for getting your work included in Emacs that I'm not sure you have gone through yet. If you have, I apologize. You should go to https://savannah.gnu.org, if you are not already a member there, then sign up and login. Then go to the Emacs page, then click on the members list. There is a link there to a form for requesting inclusion on the project. We won't need to have an assignment on file until you go through this process and have your work included. Putting your work in the public domain is okay, as such work can be included in the GNU Project. The only issue really is that proprietary developers can use your work if it is in the public domain. As for the rest of your questions, I think we should just hold off on them until we know that the work is going to be included. It's better to have someone take a look at a work before determining whether an assignment is needed, and that takes a bit of time to do. If the work is not included in Emacs, then we can about other options you might use to get your work distributed freely. Thanks again for your interest in contributing, and good luck. > [jasonspiro3(a)gmail.com - Sun Nov 30 20:41:10 2008]: -- Sincerely, Donald R. Robertson, III, J.D. Assignment Administrator Free Software Foundation 51 Franklin Street, Fifth Floor Boston, MA 02110 Phone +1-617-542-5942 Fax +1-617-542-2652 ---------- From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Mon, Dec 1, 2008 at 11:57 PM To: copyright-clerk(a)fsf.org Cc: implementations-list(a)lists.ourproject.org Hi Mr. Robertson, No, I have not applied to get either of the two works into Emacs yet. I and others agree it that Vimpulse is too buggy to get into Emacs yet. And I don't care enough about viper-in-more-modes to try to get it into Emacs. But I want to make sure the projects can easily get into Emacs when needed. So I think I will get all the contributors to release their contributions to the public domain. I will probably do so by asking them to go through the clickthrough public domain dedication process at http://creativecommons.org/license/publicdomain-2 and then forward the email receipt they get to the bug tracker. Do you think that is a good and safe method for people to dedicate work to the public domain, that will work no matter what country they come from (e.g. Canada, Italy, China)? If you reply, please make sure to Reply to All. -Jason ---------- From: Donald R Robertson III via RT <copyright-clerk(a)fsf.org> Date: Mon, Jan 26, 2009 at 7:43 PM To: jasonspiro3(a)gmail.com Hello, Thanks again for your interest. Sorry about the delay, your previous reply some how slipped past me. I can't really recommend any particular course of action in how to put your work in the public domain. In any event, even if all the contributors place their work in the public domain, we would still need at least a disclaimer from each of them if the project were to become a part of emacs. So I'm not sure how much time it would actually save to have them do that instead of just assigning their work when the time comes. Thanks again for your interest, and good luck with your project. > [jasonspiro3(a)gmail.com - Sun Jan 25 13:44:44 2009]: > > Hi Mr. Robertson, > > I still didn't get a reply yet, so I am resending my message to you. -- ---------- From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Tue, Jan 27, 2009 at 7:15 PM To: "Vi/Vim implementations: developer discussion and user tech support" <implementations-list(a)lists.ourproject.org> ---------- From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Thu, Jan 29, 2009 at 1:15 AM To: copyright-clerk(a)fsf.org Cc: "Vi/Vim implementations: developer discussion and user tech support" <implementations-list(a)lists.ourproject.org> On Mon, Jan 26, 2009 at 2:43 PM, Donald R Robertson III via RT ... Thanks Mr. Robertson. But when the time comes: 1. What if I can no longer reach them? 2. What if they then no longer want to bother with copyright assignment? I suspect it'd be very hard to then remove their code from Vimpulse. Are there any other options? For example, can you contact all contributors and ask them to assign copyright now, even though it's not strictly needed yet? Or can you send me a PDF of the FSF copyright assignment contract? I could then get every contributor to sign it and fax it to me when I merge their code. Then, when the time comes, I will fax my pile of contracts to you. When you reply to this email, please click the "Reply to All" button so that your reply will go to the whole Vimpulse mailing list will get your reply. Otherwise, I plan to forward all your future replies to the mailing list manually. If this is not ok, just let me know. ---------- From: Donald R Robertson III via RT <copyright-clerk(a)fsf.org> Date: Fri, Jan 30, 2009 at 7:28 PM To: jasonspiro3(a)gmail.com Hello, Thanks again for your interest. The FSF does not accept assignment contracts from the wild, so we don't provide PDFs of those forms. Although it is possible to have everyone assign their work now, I believe it is our policy to not accept assignments when it isn't clear that the work is becoming a part of GNU. In general, even when it is just an individual seeking to assign work and they haven't contacted the maintainer to see about getting their work accepted, I tell them to hold-off on the assignment. Sometimes, if a contributor's work is not copyright significant, we don't need an assignment from them. Other times, life gets in the way and the person never ends up contributing. In any event, it is usually the maintainer of the project who sends someone my way looking for an assignment. If you guys haven't already signed up at https://savannah.gnu.org, then you definitely should. You should talk to them about your project (I'm sure they'd be interested in it!) and about your desire to have it included in emacs someday. Even if the work isn't ready for inclusion yet, they might see it as a promising possibility. That's really the first step and the best step you can take in getting your work assigned to the FSF. You'll have to contact them anyways at some point in the process, so you might as well get it out of the way. Thanks again for your interest, and good luck with your project. ---------- From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Sun, Feb 1, 2009 at 6:51 AM To: copyright-clerk(a)fsf.org Hi, I emailed emacs-devel about getting vimpulse into emacs, as you suggested. The subject line of the mail I sent is "What to do so Vimpulse (Vim emulation plugin) can get into Emacs?". If I may ask -- why not? If I may ask this too -- why not? OK; how many lines of significant code is usually required for a contributor's work to be copyright significant? I have contacted them, but through the mailing list, not through Savannah, since the form you mentioned seems not to be the way to contact them all. Cheers, ---------- From: Donald R Robertson III via RT <copyright-clerk(a)fsf.org> Date: Tue, Feb 3, 2009 at 8:39 PM To: jasonspiro3(a)gmail.com We receive hundreds of assignments each year, and the easiest way to ensure that the forms we accept match the current language we are using is to have the forms printed on our letter head. If we just accepted forms from the wild, we would have to check each one to make sure that the language hasn't been altered or outdated. As a non-profit, we don't have the resources to go through such a process. I'm not really sure, but, for one thing, the costs associated with just getting the paperwork together and storing it is not zero. There might also be issues with whether we want to take on any responsibilities in regards to code that isn't becoming a part of the GNU project. My guess would be that getting into GNU is sort of a gate-keeper to ensure that we only expend resources on taking ownership on things that are going to be beneficial. It's not really an exact science. Copyright protects a somewhat nebulous idea known as expression, and that concept isn't necessarily tied to length of a work. A very short work, like say a poem, may contain copyrightable expression whereas a very long work, like a phone book, may not. A sort of rule of thumb that we use is that something less than 15 lines is unlikely to be copyright significant, while something more than 15 lines is more likely. In the end, the maintainer really has to look at the quality as well as quantity of the work and decide whether it's worth assigning or not. ---------- From: Jason Spiro <jasonspiro3(a)gmail.com> Date: Mon, Feb 9, 2009 at 6:17 AM To: copyright-clerk(a)fsf.org Mr. Robertson, thanks for all the time you have spent answering my questions. As you suggested, I asked the appropriate folk if Vimpulse can be accepted into Emacs. But according to what they answered[1], I will need to find a new maintainer first. I will let you know once I have found one. ^ [1]. http://thread.gmane.org/gmane.emacs.devel/108538

1 0