- Inkscape Board of Directors

Developer Meeting March 10th 2023
by doctormo＠gmail.com 11 Mar '23

11 Mar '23

Hi Everyone, Another week, another developer meeting. Thank you to everyone who responded with enthusiasm for having these notes available. I'm going to see how long I can continue to write them for y'all. Present: Tav, PBS, Martin, Mikekov, Jabier, Paopre, Jonathan, René, Vaibhav # Hackfest The hackfest is in a few days! Most people will arrive early on Monday, two members are arriving late in the evening and will have to book a hotel for the night as there's pretty tight restrictions on arrival at the venue. René will be giving further details about meals and other details, check the wiki for full details. This goes for leaving the venue too. We will be expected to be entirely gone by Friday and 6pm. Several people will be making their way to the city to hang out, or book into their next hotels if they're leaving at a later date. We have a bunch of ideas of things to do during the hackfest. Several people will probably want to conference seperately or hack in their hotel rooms if there's too much activity in the conference room. # Release Alpha, is held up by a few merge requests, we're trying to unclog the delay. - Marc wasn't present, but we have some questions to ask about how the release process is going. - René has everything in thanks to Mikekov - Jabier has two merge requests, we reviewed them. A Drag and drop has been tested but should be defered as it's not high priority. - PBS has a few issues he wants to fix before the alpha is released, One is an async rendering issue with rectangles not being selectable immediately after creation. The other is a missing library on windows called 'libangle' detected by lb90. - Reverting the Undo changes is critical! We looked at a reported crash together in the meeting, but this seems unrelated to the reversion. - The opengl performance on default Gtk on X11 can go down to 1fps with serious issues. It's a problem in Gtk and upstream gives approval for simply forcing gtk to use the image backend renderer. Questions were asked about what was the right course of action. # Other Items - Martin has refactored the export dialog previews to not block on loading large files. Lots of improvements to the export dialog. - The font dialog has a small issue with folders in the font dialog being created, it's creating a system folder when it shouldn't be. - Nico has made good progress with the ai importer this week. He's working about 20 hours a week and progress is steady. - Paopre asked about his proposal to reorganise the test suites so each of the unit tests, integration tests, and others into more formal and defined directories. Advised to document the proposal and present it to the developer team so we can have time to look over it. # Group Activity This week we skipped doing the merge request review and instead did some code review. Walking through the changes involved in the export dialog changes. See https://gitlab.com/inkscape/inkscape/-/merge_requests/5148 for details. Thanks everyone for coming. The next scheduled meeting will be after the hackfest, but we'll likely have one at the end of the hackfest instead. Keep up to date with the calendar at https://inkscape.org/cals/event/1/https://inkscape.org/cals/event/1/ Best Regards, Martin Owens Inkscape Developer Team

1 0

Notice of resignation from the Inkscape PLC (AKA Inkscape Board)
by C R 07 Mar '23

07 Mar '23

Hey folks. Sorry to offer up two resignations in two days, but I've been losing sleep and realised there are some things which have been bothering me for quite a long time now. I think it boils down to this: My visions for Inkscape simply are not in-line with most of the rest of the project, and this is why I'm offering up my resignation. I firmly believe that PLC seats should be reserved for people who can represent the project's interests, which are decided by those who do the lion's share of contribution. I do not think it's appropriate for people to vote on funding for parts of the project they have no hand in, and as such, I can not bring myself to continue voting on funding things for development, as I'm not a developer. The truth is, that I believe the development team should have final say in the project on this matter, thus my vote would always be "yes" for anything the developers have collectively decided. The same goes for the Vectors, Testers, and any other group within Inkscape - they decide they need funding for x-thing they handle and are responsible for in their part of the project. It's a 100% yes from me, always. I think that's what a truly representative PLC should look like and act like. I do not think that the Vectors should be able to downvote development funds, just as I don't think the developers should be able to downvote vectors funds for the oureach activities that they handle. If a group does not do a thing, considered important for the project, they simply should not vote on those things either. Recently I've seen an increase in attempts to build some processes to ensure "fairness" in funding. While in principle, this sounds like a good idea, I believe it's achieving the opposite. I think it's incredibly unfair to punish those who have and do contribute constantly to inkscape, because of supposed "conflicts of interest". This was made crystal clear to me recently when I requested funding to make a video under the same terms as last year (not setting any precedent for next year). If it's a conflict of interest for me to request funds, and no one else steps up to request funds on behalf of my video project, I'm stuck in a catch 22 situation, and the video just doesn't get funding. Maybe that's just lack of interest in a video from the Vector's side, but then why should they be in charge of voting down something as essential to outreach as a release video? Let me make it clear - I don't care if it's me or someone else who does the video. If there were someone else with the capability to do it in FOSS to a professional standard at this late hour for this release, I'd just hand it over to them, since me doing it is apparently so contentious within the project. What upsets me is people seem to be voting for what they think the bureaucracy should look like, and not for what would benefit the project most at this time, while organisational efforts are incomplete. I fully understand wanting to put a fair process in place, but not a finger has been lifted all year to do that, and I'm tired of fighting endlessly with people's personal sense of morality, while the project's development and exposure suffers. People seem to want a rigid, inflexible (and entirely undeveloped) process for doing things, and are willing to toss important deliverables for the project under the bus to get their own way. I've loved some of the progress we've made together in getting things funded, but I can't fight everyone on these things and still consider myself a good project representative. I therefore resign, effective immediately. Thanks for your support, friendship and understanding. Hope this isn't too much of a downer for folks. I'm honestly feeling better already. :) For Inkscape, always. -C

4 5

Need new Inkscape project head moderator
by C R 06 Mar '23

06 Mar '23

I'm stepping down as head moderator. Someone else will need to do it. FYI -C

2 2

March 2nd Meeting notes
by Martin Owens 02 Mar '23

02 Mar '23

Developer meeting for March 2nd, Next meeting is March 10th https://inkscape.org/cals/event/1/ Big thanks to everyone who came to this week's developer meeting. Attending: Marc, Jonathan, Martin, NPJ, René, Mikekov, Nicco # Hackfest The hackfest is a week and a half away! Pono is coming to the hackfest, this will be good and it'll be great to see him in person. SFC will pay in advance for the hackfest using an international wire transfer. The money will need to be there by next week if we want this all to work out. But this is good news that there are no further discussions needed to sort this out. Reviews of the wiki as well as the developer introduction page are planned activities. Especially to help GSoC students with their common questions. Marc is planning on working on dithering at the hackfest, other developer have various activities planned for their time at the hackfest. There's no set developer meeting during the hackfest, but every day will be a meeting. Streaming options and interviews are also planned. # Releases and Release Planning Marc has been busy but is planning on releasing an Alpha version this weekend. He wants to check on the windows store release, but is confident that windows builds can be created. René has a specific set of fixes that need to make it into the release because all builds are broken for macOS at the moment. The alpha will wait for this merge request to be complete, and is expected to be complete later today. The openmp library update fixes the latest macOS specific crashes which was only found because someone opened the about screen. Marc needs help uploading things to the Ubuntu PPA regarding cpack. May need to reach out to Ubuntu maintainers to get the process working well. René will remove dithering support in the macOS release because it's not stable and is causing problems. He will also be making an ARM build of the macOS package using our own CI builder infrastructure which should mean we have these builds for alpha. Mikekov has a few merge requests that need attention and wanted clarification on what is acceptable to merge during freeze. At this point we wouldn't stop changes to strings, but we don't want large untestable changes, so most additions will be made on a case by cases basis, but the size of mikekov's changes should be ok. # Importer Project Nicco, Jonathan and Martin did an on-boarding session on Monday, and Jonathan has been working with Nicco all week to guide the creation of the new extension. There's work done on the environment and the testing framework to make sure the entire process is tested and that it fits in with how our existing extensions work. Nicco has been looking into the format specification and putting together the priority for what pieces should be worked on first. Basic shapes are expected to be working by the hackfest time frame. The type of python dialect/parsing programming which is not-object-orientated is interesting to Nicco. Waiting on getting a message out through the Vectors team. # Bug Accelerator Program Martin has published the first blog post detailing the work done for the program last month (Feb). This is the tail end of the month and Rafael will become available this month (March) Other developers are happy with the progress. Instructions have been made on how to encourage and prioritize bugs. Please use the Inkscape 1.3 tag, and increase the priority on issues which should be fixed first. # Gtk4 Upgrade Proposals Jonathan is going to bring the proposal to the hackfest so it can be worked on in person. Mikekov is concerned about the instability and lack of commit access that will be available during the upgrade. Reassurances have been made that the upgrade would not take years to complete and this is one of the reasons we want to commit paid resources to pushing it through at a quicker pace, so we don't block valuable contributions that would otherwise go to waste while Gtk4 upgrades were proceeding. Martin recommends doing a risk assessment at the point where all Gtk3 compatible actions have been completed and the fork to a gtk4 branch is needed. Everyone agrees that Inkscape's main branch should never be in a state where it doesn't compile. Although we might want to freeze work on the UI folder during the upgrade process. # Vectors There's some discussion that the vectors team doesn't have the resources it needs to respond to requests from the developer team and we wonder how we can help. Maybe we need to do more recruitment, or do join activities to raise excitement. It's hard to know what might help. René proposed that the developer team may want to request the production of a release video if there are no objects from the vectors team. Although Martin was more cautious and wanted to instead see if the vectors team could be encouraged or resourced rather than circumvented. This will need to be an ongoing conversation with the members of the vectors team. # Other work - Martin has been working on speeding up the rendering pipeline, specifically the way bounding boxes are calculated over and over again. - Jonathan did a pdfimport frenzy to find issues and ended up opening 11 new bug reports. This might indicate we need more testers for 1.3 to improve our coverage. - Marc wants the old-files text fix to be tested more rigorously before it's accepted. - René says the performance of test running is getting worse and we're not sure why. - NPJ reports that macOS is very fast and wants to congratulate us all on a job well done. # Merge request review We looked at four merge requests. One was given back to the developer (mikekov) and three others were moving code around by new contributors who never go back with comments and so were closed. Thanks everyone! See you all next week.

1 0

VOTE: Funding for Bug Mugs
by C R 01 Mar '23

01 Mar '23

Hey folks! As we wrap up the Bug Migration Game, we'd like to reward our bug movers and people involved with making/running/promoting the game with a bug mug. I'd like approval for (up to) £200 in funds for sending these out to people, though it may wind up being less depending on what deals I can get. This includes cost of ordering them and shipping them out to people. I've attached a mockup of the bug mugs, in case you haven't seen them yet: The vote is for up to £200 in mug and shipping costs. Vote is for 1 week, as is tradition, but would appreciate votes before that to help move things along. Yay: Fund bug mugs for volunteers Nay: Do not fund bug mugs for volunteers Thanks!

7 10

Developer Meeting notes for Feb 22nd 2023
by Martin Owens 22 Feb '23

22 Feb '23

Attending: Marc, Martin, NPJ, Nicco, Johnathan, René, pbs, Mikekov The next developer meeting is 2nd March 2023: https://inkscape.org/cals/event/1/ ---- How the bug accelerator program is going (Martin) Bug list for 1.3 is being used https://gitlab.com/groups/inkscape/-/milestones/11 developer should add the milestone 1.3 if they want issues to be in the mix. The advantage is that this list provides a sorted and updated list and so the blocker issue has been closed and instead people should mark issues as blocker using the tag. Lots of bugs fixed and lots of merge requests are now open waiting for review. When Rafael becomes available merge requests will have a reviewer to move code faster. On the effectiveness side; Martin says other paid contracts have been re-scheduled so time is available to work on Inkscape issues, so it's functioning as intended to create time for inkscape at this early stage. Though this is a self-reported data point. A new blog is available to contributors. This will be used to report on work done in the bug accelerator program, but can be used by other contributors if they want to talk about their work and have it available on the inkscape website. Please contact the web admin for access. ---- The recent Cairo rendering failure is related to Marc's dithering patch. It was suspected that it was just related to using an updated version of cairo, but the patch itself is implicated and Marc things it's becuse of the way the dithering code throws an error which pbs says will cause the whole rendering stack to then reject any other rendering instructions. ---- A Pango problem that has appeared is an upstream issue with the fontconfig backend. To fix this the version is restricted to 1.50.11 for all builds and the windows build is pinned to 1.50.11 for now. The issue has been reported to upstream although they say they can't guarentee a fix. ---- A version suffix was added to inkscape's library file (*.so, *.dll etc) this allows the library file to be placed in predictable places and should make it easier to create C++ extensions which can link against the Inkscape library of a specific version despite not having a stable API. Getting these library files into the right system directories to make them useful is a possible task, as well as providing header files or a -dev type package to allow developers to work without building the entire of inkscape first. - This causes a problem for macOS builds that didn't expect the version to be there. but René is working on a packaging fix. ---- Mikekov is working on a few features and improvements, mostly UI related fixes. A new milestone has been created by Marc so anything that we don't expect to make it into 1.3 can be tagged for the next release and not merged by accident before the branching date. Marc asks if the font collections feature is ready? It needs to be rebuilt, but looks to have problems and may need to be posponed. The original developer may also not be available. --- What kind of plan should we have for Gtk4 migration? After the 1.3 branch (or release) we will allow much more riskier code merges that progress us towards Gtk4 support. So master is expected to break more often, but we want to make sure Inkscape compiles so any changes will still be expected to compile even if they cause large bugs. Having this high risk development policy after the release is most ideal and sets expectations in the whole developer team about what we need to be focusing on. Jonathon is creating a paid contract to hiring people to work on the Gtk4 transition, please join him on rocket chat to look at the proposal draft and see what kind of paid work we need to ask for in order to push hard on getting Gtk4 transition done with the smallest amount of disruption it's possible to have in our development cycle. ---- We were concerned that Nicco had not recieved the contract for the AI importer work. But Nicco was available at the meeting and has recieved the contract. Next is signing it and sending back to the SFC, and Johnathan is organising a date to do onboarding with Martin. ---- The Hackfest is coming up very shortly. There's ongoing discussions with the SFC and René regarding paying for the venue costs upfront so the particpants won't have to pay the large fees that will be due at the end. We are told Karren was unavailable so there may be some delays reviewing the venue contract and sorting out the payment. Though we do want to make sure this happens because we don't have a lot of time now. ---- We are accepted into the GSoC program, number of applicants isn't set yet. You may advertise about our particpation in GSoc 2023, and please make sure to update the list of projects on the wiki https://wiki.inkscape.org/wiki/Google_Summer_of_Code ---- We looked at three of the oldest merge requests as part of our regular end of meeting activity: One was assigned to Adam (sorry adam!), two were closed as out of date. Thanks for coming everyone! Best Regards, Martin Owens

1 0

URGENT VOTE: INKSCAPE STICKERS & BUSINESS CARDS for Hackfest (and other events).
by C R 17 Feb '23

17 Feb '23

Hey everyone! We need moar stickers! These are both for the hackfest stickers, and for the stickers for SFC and Inkscape reps attending conferences to hand out. We'd also like another set of business cards with the hotkeys on the back. The SFC will be paying for these directly and shipping them out to Rene to distribute at the hackfest, so we need votes asap! 500 Cards: £44.47 2000 Inkscape Stickers: £133.40 1000 Hackfest Stickers: £30.00 Shipping: £30 Total: £237.87 Additional shipping may be required to send out hackfest stickers, so approval for up to £100.00 extra to reimburse for shipping stuff out to folks. Total incl shipping (up to) £337.87 Vote: A. Fund stickers and cards B. Do not fund stickers and cards Thanks! -C

7 19

VOTE: Mastodon.art funding
by Ted Gould 16 Feb '23

16 Feb '23

Background Inkscape currently has a Mastodon account on the mastodon.art server. It is an art focused Mastodon instance that has a wide variety of artists, but also allows anyone on the greater Fediverse to see the content Inkscape posts. The service is run by a set of volunteers that have a Patreon-style set-up to cover their costs of hosting, and other things that the project needs. This is their only source of funding (they don't sell ads, etc). Many users contribute to that as the "cost" of having an account on the service. Inkscape has not previously contributed to this service. There is unfortunately very little precedent on how much a social network account should cost. Most major networks are venture or ad supported and give out the accounts for zero-cost. Though it seems likely in the future Twitter will be charging $8/mo if you want your posts to be seen. No one is distinguishing between user or project account, though it is likely their costs are different based on volume. The mastodon.art service has a few ways to contribute, I'll work with them to figure out which is best to ensure they can use the funds at the lowest fee and is easy for the SFC to work with. All amounts are monthly for the 2023 calendar year. Marking each of the votes with "or less" and I'll take the highest amount that gets a simple majority. Please use the "Other" option if that's not how you'd like it to work. Proposal Vote to fund Mastondon.art for the following amount monthly for the 2023 calendar year a. $5/mo for a total of $60 for 2023 b. $10/mo for a total of $120 for 2023, or less c. $20/mo for a total of $240 for 2023, or less d. $__/mo for a total of $___ for 2023, or less e. None f. Other: ____ Timeline Proposed: 2022-12-11 Closes: 2022-12-18

3 4

VOTE: Mac Hardware reimbursement for RdH
by C R 14 Feb '23

14 Feb '23

RdH has requested reimbursement for the project to pay for his new Mac Mini, which is necessary for him to develop Inkscape's Arm chip releases. This is crucial to MacOS development on their desktops and laptops now and into the future. Please vote: 1. Reimburse RdH the entire amount: 2,584 € 2. Something else. -C

4 7

VOTE: OSU OSL Donation
by Ted Gould 06 Feb '23

06 Feb '23

Background For our webserver and other online services we rely on OSU's Open Source Lab to run the physical servers. They offer this for free to Open Source projects, but in the past we've made donations when we could to help support their project. For example: * https://alpha.inkscape.org/board/referendums/resolutions/2014_osuosl_donati… * https://alpha.inkscape.org/board/referendums/resolutions/2015_osuosl_donati… In the past, we've donated $200. We've grown. Inflation has happened. That is probably not the amount we'd want to give today. Martin reported that the webserver alone is a decent sized machine: https://alpha.inkscape.org/board/meetings/transcripts/2023-01-06.log.html#t… And if you look at the Linode calculator I think if we just used Linode that would cost ~$20-$30/mo. https://cloud-estimator.linode.com/s/ I've structured the voting similar to how it was done for the Mastodon donation vote so that hopefully provides flexibility for folks without requiring multiple votes. Pono has said that the SFC can handle the contribution to OSU OSL along with other donations from SFC projects, so he will execute the donation for us. https://alpha.inkscape.org/board/meetings/transcripts/2023-01-06.log.html#t… Proposal Donate the following amount to OSU OSL: - a. equiv. $20/mo for a total of $240 - b. equiv. $30/mo for a total of $360, or less - c. equiv. $40/mo for a total of $480, or less - d. $__/mo for a total of $___ , or less - e. None - f. Other: ____ Timeline Proposed: 2023-01-11 Closes: 2023-01-18 Ted

7 9