Next major version, diaspora*

(Flaburgan) #1

A new major version of diaspora* is released when there are enough major changes in the develop branch. It used to be twice per year, and now it’s more about once per year, usually around the 27th of August, date of the switch of the project from the founders to the community.

This date is now in less that 2 months. I’m opening this topic to start to think about it, and try to see if we will have enough changes to release a new major version at this date. To anticipate a bit would be helpful both for dev and communication, and could maybe even lead to call for people to help, like students who should have more time during the summer.

The current changelog already contains the two new features:

  • Add client-side cropping of profile image uploads #7581
  • Add client-site rescaling of post images if they exceed the maximum possible size #7734

The current features PR looks advanced enough so the release can wait for them to be finished:

Am I missing things here, are you aware of other “user visible” changes which could ship during the summer?

Even with those, the changelog seems too small for a major version in my opinion. It would be nice to have a few more big features to deliver. Maybe bringing back likes on comment, as the federation part is done, and the backend part is partially there?

What do you have in mind? How do you feel about this release? Should we try to target the 27th of August even if the changelog isn’t really big because we should stick to at least one major release per year, or do you feel that we should wait to have a big changelog, even if that delay us until who knows when? Do we want to try to boost the development during the summer, maybe by organizing an online hackathon or something?

(LoYgg) #2

Hi, does the «Archive import backend» means the migration from one pod to another (with all contacts, aspects and posts of the migrated account) will be effective and smoothly working ?

If yes, it seems to me that it is a hell of a feature, easily big enough to justify a major version. If I remember well, it was one of the key feature planned at the beginning of the diaspora project by the founders, but never been achieved yet. It talks about freedom (to move from one server to another) and your own data handling, which are at the core of diaspora’s heart, aren’t they ?

(Flaburgan) #3

It definitely is, and we’re super excited to see it finally live, thanks to @comradesenya for his work and @supertux88 for the review.

However, if you’re perfectly fine with the pod you’re on, then this new feature doesn’t bring you anything…

(Dennis Schubert) #4

No. It means that the backend supports it. We have no interface for that, and the PR will not contain an UI.

Anyway, back to topic. We will not have a fancy stream this year, for sure. And in my opinion, we don’t have anything to show off right now, so pushing a major release feels kinda wrong. I usually also don’t like looking at what might be done, but rather at what is done. If, for some miracle, it turns out we have the migration backend worked out and have a nice interface for that, sure, let’s push a release. But as of right now, I don’t think we can get even close to targeting these tasks to any specific date.

I also don’t think we should have “one major release per year”, ust to release something. We have regular releases for the important stuff, so we should make our majors somewhat interesting.

(Comrade Senya) #5

That’s being said, I think the frontend is the easiest part and won’t take that long as all other parts of the migration feature

(Comrade Senya) #6

I wonder whether @SuperTux88 is still motivated reviewing 7660? I know it’s very hard and requires a lot of efforts and energy to do that and the job is not fun, so I won’t be surprised if he simply don’t have enough motivation and interest. I wonder if I can do anything to improve the state of things. Maybe I can split the code somehow so that it can be reviewed in batches? Maybe some online work session with realtime involvement will make it less boring to work on?

Unfortunately we don’t have a lot of people who are willing to spend a lot of time on reviewing contributions to choose from.

If we had a at least a part time paid developer with push access to the main repository, then we could have more predictable time for reviewing and merging 3d party contributions like mine. But for that we need a constant contributor who’s willing to work for money and also we need money and seems like we’re missing both. So if there is anything we can do about it is try to attract some new contributors who might at some point want to become constant maintainers.

In your opinion how do you think, would it help if we get together somewhere physically for some period of time, say, 1 week, and work on diaspora? 1 week is not much for software development, but still we might have more effective communication and solve some questions faster and make some decisions which take forever in written form to agree on. Of course that requires some planning and people have to take a week off their jobs, so the question is does it worth the troubles.

(Benjamin Neff) #7

Yes, I just didn’t have the time yet (I hoped I have time this weekend, but didn’t). It’s still on my todo list, but I’m a bit busy with other things at the moment.

Review together (via IRC or something with voice like mumble) to clear questions immediately would be cool, but I don’t know if that would make the situation even more complicated, because we would need to find a common date :wink:

(Comrade Senya) #8

Just ping me when you decide about your plans, and I’ll tell you if
I’m available or not.

(Flaburgan) #9

To meet IRL would of course be the best but that’s time (you need to travel etc) and money consuming. Audio meeting seems a nice improvement. I’d be glad to join too. Tell me when you two picked a date and I’ll see if I’m available.

(Rob) #10

Any updates on release of the next version? Is there already a changelog/extended feature list? Any known bugs list that still need to be resolved before release?

(Flaburgan) #11

Hi rob, you can see the changelog of the future versions on github. There is no blockers for the coming release at the moment, it’s just not released yet because we feel the changelog is not big enough unfortunately. So if you want it to happens, open pull requests for features you love! :smiley:

(Rob) #12

Hi @flaburgan
Thank you for the reply and the call for PR’s.
I can do a lot, but unfortunately I am not a coder. My merits are more towards operational support (admin stuff) and community support (community management) and I will try to help out on those parts of the project as much as possible.

Adding feature requests is always an opytion ofcourse, but IMO features should be backed by the community as being necessary or extremely usefull. I don’t think it is wise to drop feature requests on Github without having discussed such a feature in the D* community first.

(Hank G) #13

I know I’m late (and still a newbie) but adding two factor and archive import are huge features which I think would warrant a major release push if the software process here is that major changes need to go out with major version updates. If those aren’t done and tested then I wouldn’t bump the release just to say we bumped the release in the year. However once those are done I’d say it’d be time. I love both of those features and can see using them as soon as my pod has it integrated. I like my pod but want to create my own. However I don’t want to start my account over from scratch either. Anything on the internet I prefer to have two factor authentication too so this would make me more comfortable as well.

(Hank G) #14

Now that I’m back on the internet/social media this is high on my priority list for filling my free time :slight_smile:

(Hank G) #15

On the “should we do” would the REST API feature need to be integrated in a major point release? If so where is the development timeline on that? If it is somewhat near the status of these features and would need to wait for a full point release then I’d say hold off on that. However if that is somewhere where we could have a major point release just for that alone within a few months of then I guess it’s moot.

(Flaburgan) #16

As far as I know, no progress is done on the API side at the moment, so it should not be a blocker.
And for the question “should we release for the birthday” well, it’s past so obviously the answer is no :stuck_out_tongue:

(Hank G) #17

Yeah I could tell the discussion about the actual release date had come and gone but the question of whether these were sufficient features to warrant the number bump seemed still relevant.

(Flaburgan) #18

As there is very few progress on those topics, I don’t know. Maybe someone will step up and the API will be done before the import feature? Who knows…

(Hank G) #19

It’s something I’m hoping to work on actually. I’m not making that my warm up rounds though. I’m looking to do smaller surgical things, preferably that would expose me to both back end and front end, before tackling something larger like that.

(Hank G) #20

This topic was created in July with the idea of an imminent major release in August 2018 which there weren’t enough features for. If we decide to keep with the August release tempo for the major release, which I think would be a good idea, then thinking ahead to the next several months of development for what we could target for August 2019 would be a good thing. Now that the API is getting ready to go into its next stage of development I am looking at what I think could be a good set of major features for, based on the above but with some additional ones. These are things that I as user would really like to have, that I’d be willing to work on, and that I think are feasible to do in a 6-7 month time frame. Here are my thoughts for what I’d like to accomplish for and what I’d be willing to sign up to work towards. Thoughts? I can make this a new thread if that’s preferred.

  • API in production (almost staged for PR)
  • Posts
    • Liking Comments
    • Disabling Comments
    • Closing polls
    • Editing posts
    • Editing Comments (if protocol already supports that as well)
  • Photos
    • Streaming instead of greedy load
    • Pasting directly into posts (already staged for PR)
  • Mobile
    • Aspects multiselect
    • Autocomplete names when posting/commenting
  • Archive Import (already staged for PR)
  • Two Factor authentication (already staged for PR)
  • Advanced search and custom streams