Extension update policy

[bmarkus]

In the last few weeks I reserved some time to go through my extensions and to update them or recompile for TC 3.x making new build scripts eliminating manual fine tuning specially for older builds. It requires time, checking upstreamn sites, bug fixing when it can't be compiled on 3.x, adding icon, correcting TC specific files, etc.

I was really suprised to see that certain extensions simply just replaced with a new build without telling me something was wrong with them, do I want to keep maintenance ownership or not, etc. Such extensions are moc, windowlab, etc.

Once upon a time there was a policy not to hijack ownership on extensions. Sorry to see it is gone now. I like TC, but do not have time to waste and loose what I contributed to TC. I do not know, which extension will be the next or if I submit a new when this will be replaced without notice.

Now I stop my TC activity for a while. There are so many other things to do. Please feel free to do anything with extensions built by me.

Good bye.

[SamK]

This is indeed bad news.

You have been one of the most active and supportive community members.  TC will be poorer without you.

Many thanks for the advice you have offered me and for your extensions which I have gratefully used.
 

[Juanito]

Once upon a time there was a policy not to hijack ownership on extensions. Sorry to see it is gone now. I like TC, but do not have time to waste and loose what I contributed to TC. I do not know, which extension will be the next or if I submit a new when this will be replaced without notice.

Jason will probably be able to answer better than me, but the idea is still that extensions should not be hijacked without asking the owner's permission via a forum post or personal message

[tinypoodle]

Purely coincidentally I happen(ed) to do some research on forum and wiki these days about creating of extensions, about technical aspects as well as policies.

Following thread "Extension maintainership" - marked as sticky topic - seems pretty clear to me:
http://forum.tinycorelinux.net/index.php?topic=5287.0

The fact that this thread has only one post allows for the interpretation that the subject is undisputed (so far), anyone disagreeing could just have done so by replying with a post.

Comparing the 2 versions of the 2 extensions mentioned:

windowlab appears to be an update without doubt, containing reference to former version.

moc-svn slightly differs in name from moc. Having noted that in more than one instance more than one version of the same app coexist in repository, one hypothetical explanation could be that moc-svn was never meant to replace moc, but rather to be an additional extension.

Noteably date for the moc-svn in http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/tcz_3x.html is absent (unusual).

http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/3.x/tcz/moc-svn.tcz.info has an unusually long "Description" entry, part of which would rather be suitable to fit under "Comments:".
"Current:" entry is of unusual (compared with other extensions) date/time entry - which I could imagine to be the cause of the lacking date on extensions page (as in link above).

bmarkus:
As by my interpretation you do not seem to dispute existing policies, but rather instances of their practical implementation, I would propose that before you draw any final conclusions and make any definitive decisions you would wait a bit for possible replies/reactions.
Miscommunication and miscoordination can happen in a project with a structure as TC has (user contributed content, volunteering).
But also I would like to believe, that once miscommunication or miscoordination does happen, that issues could be resolved with a spirit of fairness and broader consent, under the condition that there is good faith from all which are concerned/involved. Should a spirit of antagonism rather than collaboration prevail, then that would be most regretful.
(This does not imply that I would not understand your personal disappointment).

Regardless of what your final decision on this is, thank you for your countless contributions up til today (your extensions as well as your forum input).

[aus9]

bmarkus

You were my inspiration to look at TC from the LXDE forum we know about.

If you go....I wish you luck.

regards

gordy

[curaga]

Sorry to see you go.

The policy is still up, but as written there, Jason can't check every submission. It is assumed there is permission.
Did you contact anyone on the team when you noticed this?

[roberts]

I am distressed to learn of this and the seemingly swift and drastic response to leave. Bela, your help in providing "use cases" and early testing of both recursion and freedesktop implementations help Tiny Core achieve these milestones. Your contributions in the forums and extensions has helped many enjoy expanded capabilities of Tiny Core. Hope you will reconsider your decision.

On the other hand, I feel compelled to state the following. Team Tiny Core is all volunteer. Each gives of their time, interests, and talents to make Tiny Core the best possible. We all try our best. I am positive that no ill will was intended. Everyone needs to realize that errors, omissions, and issues will always occur in both software development and in coordinating software submissions from a community. While we do post rules for forum use, remasters, and extension submissions. We can only hope that such will be honored. When not, it is usually best to notify a team member of any concerns. It goes without saying that it is always best to try to achieve amicable resolutions. Such is paramount for success in all endeavors. 

[jur]

I am distressed to learn of this and the seemingly swift and drastic response to leave. Bela, your help in providing "use cases" and early testing of both recursion and freedesktop implementations help Tiny Core achieve these milestones. Your contributions in the forums and extensions has helped many enjoy expanded capabilities of Tiny Core. Hope you will reconsider your decision.

+1!

[Jason W]

The issue of existing extensions getting clobbered by uninvited updates is a potential casualty of our open submission process.  With 2509 tcz extensions in the repo now, well actually 2511 when my current upload gets done, it is very possible for any of us to not realize an extension already exists or some other oversight.  And of course, there are also new submitters that may not be aware of the protocol of contacting the current extension maintainer before submitting an update.  

Up until now, I left the responsibility with the community, and I think overall that has worked well.  But as the repo gets bigger, so does the potential for these kinds of errors to grow.  I have updated the extension audit script to check for the existence of an extension in the repo and it will cough up the maintainers name along with yours as part of the process for you to see.  The two should match, or your name should at least be among those in the maintainer field of the existing extension if there are more than one. I will be checking with the same script, and I will not upload the extension unless you explain that the maintainer is ok with your update if you are not on the maintainer's list in the "Extension_by" field in the existing extension's info file for an extension that already exists in the repo.  For those that are not very familiar with all the details of extension submission, please use the audit script.  I use it always with my own extensions as it is at least a time saver, and will catch many little things that are easy to overlook.  Please don't make me play email tag for an extension update that is not your own that you have not cleared.  That will cost us both time, which is probably the scarcest resource we have here.

There are so many growing pains that can occur in such a fast moving project and a fast growing repo.  We cannot always prevent them, but we can respond to them when they occur.  

As for Bela, I don't need to state that he is one of the most respected and valued top contributors.  I hope that if he stops his TC activity for a while, it is only a short while.  

[dufus]

Don't go! (But if you want to rest a little, no one deserves it more than you.)

[gutmensch]

I am distressed to learn of this and the seemingly swift and drastic response to leave. Bela, your help in providing "use cases" and early testing of both recursion and freedesktop implementations help Tiny Core achieve these milestones. Your contributions in the forums and extensions has helped many enjoy expanded capabilities of Tiny Core. Hope you will reconsider your decision.

count me in: +2!

Come on, Bela... we even survived HAL, we can't give up now so easily! ;-)

[Lee]

+3.