465 replies to this topic

[littlesat]

BitrateViewer should also be fixed now...

 

Thanks... as it is actually you who should find this stuff...

Edited by littlesat, 24 January 2018 - 10:16.

[Huevos]

It is a shame PLi cannot provide a Developer image for the people that are contributing to PLi but don't have the resource to build the image themselves. This means it is very difficult for us to test and submit new code (even Python code) to the develop branch.

Edited by Huevos, 24 January 2018 - 10:30.

[Persian Prince]

It is a shame PLi cannot provide a Developer image for the people that are contributing to PLi but don't have the resource to build the image themselves. This means it is very difficult for us to test and submit new code (even Python code) to the develop branch.

For which STB you want a test image? I can compile from develop branch if you want

[Rob van der Does]

atm I use my old VU-Ultimo for PLi skin testing.

Be aware I would also need feeds......

Edited by Rob van der Does, 24 January 2018 - 11:24.

[Huevos]

Solo4K (with updates, feeds and plugins?). Or OS Mega.

Edited by Huevos, 24 January 2018 - 11:37.

[WanWizard]

It is a shame PLi cannot provide a Developer image for the people that are contributing to PLi but don't have the resource to build the image themselves. This means it is very difficult for us to test and submit new code (even Python code) to the develop branch.

 

It is being discussed.

 

The main problem is that as soon as we do that, a lot of people will install it (as they believe "newer is better"), and the forum will be inundated with "bug reports" for things that are simply work in progress. Or people get angry because all of a sudden things don't work anymore because of work in progress. We already see this behaviour with the RC's we release, which are clearly "pre-release test images". Yet, people rush to install them without any intention of actually testing anything.

 

One of the things I'm thinking about is to start some sort of "developer programme" to which you can subscribe, which gives you access to a developers-only forum for technical discussions, development images and the issue trackers, which we can limit to people that actually know what they are doing. But no decision has been taken about this yet.

[Persian Prince]

 

It is a shame PLi cannot provide a Developer image for the people that are contributing to PLi but don't have the resource to build the image themselves. This means it is very difficult for us to test and submit new code (even Python code) to the develop branch.

 

It is being discussed.

 

The main problem is that as soon as we do that, a lot of people will install it (as they believe "newer is better"), and the forum will be inundated with "bug reports" for things that are simply work in progress. Or people get angry because all of a sudden things don't work anymore because of work in progress. We already see this behaviour with the RC's we release, which are clearly "pre-release test images". Yet, people rush to install them without any intention of actually testing anything.

 

One of the things I'm thinking about is to start some sort of "developer programme" to which you can subscribe, which gives you access to a developers-only forum for technical discussions, development images and the issue trackers, which we can limit to people that actually know what they are doing. But no decision has been taken about this yet.

 

+11111111111111111111111111111111111111111111111

[Persian Prince]

atm I use my old VU-Ultimo for PLi skin testing.

Be aware I would also need feeds......

vuultimo is ready (develop branch): http://www.mediafire...uultimo_usb.zip

No feed sorry ...

[Rob van der Does]

Thanks, but without feeds (plugins) there's not much to test.....

[Persian Prince]

Thanks, but without feeds (plugins) there's not much to test.....

What packages exactly do you need? Maybe I can upload some of them ...

[Rob van der Does]

I need to test virtually all plugins (for that box), or at least as many as I can handle, including the sources.

I think it's best to await PLi's (decision on the) availability of dev-images.

[Huevos]

 

It is a shame PLi cannot provide a Developer image for the people that are contributing to PLi but don't have the resource to build the image themselves. This means it is very difficult for us to test and submit new code (even Python code) to the develop branch.

 

It is being discussed.

 

The main problem is that as soon as we do that, a lot of people will install it (as they believe "newer is better"), and the forum will be inundated with "bug reports" for things that are simply work in progress. Or people get angry because all of a sudden things don't work anymore because of work in progress. We already see this behaviour with the RC's we release, which are clearly "pre-release test images". Yet, people rush to install them without any intention of actually testing anything.

 

One of the things I'm thinking about is to start some sort of "developer programme" to which you can subscribe, which gives you access to a developers-only forum for technical discussions, development images and the issue trackers, which we can limit to people that actually know what they are doing. But no decision has been taken about this yet.

 

Making a Developer image public is a mistake. Needs to be given out selectively to people who are active contributors.

[littlesat]

Needs to be given out selectively to people who are active contributors.

Please help us to define....

[WanWizard]

Making a Developer image public is a mistake. Needs to be given out selectively to people who are active contributors.

 

Which is what I am suggesting.

[Rob van der Does]

 

Needs to be given out selectively to people who are active contributors.

 

Please help us to define....

 

Being an active member on the forum yourself I'm pretty sure you (and other PLi-members) know by heart whom you would want to invite from the forum-guests. And you don't need a hundred.

[WanWizard]

Sounds logical.

 

Btw, I would not also like to include contributors, but also people who want to actively test and report.

[Persian Prince]

if [ member == "gutemine" ] ; then		# Do we need to write a comment after all these years?
	log "Not a contributor?"
	exit 1
else
	echo "Thanks GOD! You just passed the first step ;)"
fi

Edited by Persian Prince, 24 January 2018 - 17:35.

[Rob van der Does]

Sounds logical.

 

Btw, I would not also like to include contributors, but also people who want to actively test and report.

The way we did it @ ViX is: create a new forum group (called beta-tester or something like that) with access to a dedicated sub-forum and the dev-images. Make an invent of the required images (not all images will be required) so the build-time and hence the build-cycle can be short (i.e. build when needed (e.g. after a lot of changes or changes in cpp's), not specifically during the night). So maybe 3 images on a day, and none on another day.

And indeed: PLi does have a rather large group of devs (compared to other teams) so IMHO your need is active testers.

A good bugtracker is indeed a prerequisite, but this can also be done via a sub-forum.

 

Invite people who caught your attention on the forum, and make some kind of a rule that no participation = end of betatester (allowing of course for people who are not 24/7 active).

Also make sure that PLi devs actively participate in discussions with testers.

[mrvica]

@gutemine was not participating in this item at all

[WanWizard]

@Rob,

 

My plan is along similar lines, but most of the build- and distribution systems are already in place. The bugtracker also already exists, but isn't populated yet.

 

Challenge in my opinion is vision, direction and coordination. Especially if you have large or more complex projects that need multiple developers over a longer period, you can't do without. Currently most activity either comes from someone thinking of something by himself, developing it, and introduce it without much discussion and without a plan the change will fit in.

 

We may have a relatively large group of members, but the number of real developers can be counted on the fingers of one hand. Easily. Ideally we could use a few more, both C and Python, so we can attack more major changes.

 

Also needed: forum moderators. Currently a lot of dev time is lost because the devs have to spend time following all forums and answering questions, time that could be spend better elsewhere.