Great to see the project moving along, but it's been a persistent feeling of mine that FOSS is starting to get left behind in the dust. The design space has been very rapidly moving towards online, live collaboration tools like Figma, Sketch, Adobe XD, etc and FOSS just has no answer for this market. Even a lot of the big name proprietary tools are finding themselves becoming rapidly obsolete in a world where people want to IM a link and just have it reflect the latest version rather than attaching files back and forth.
It's not even clear how FOSS would even work in this market or how it would benefit the user when it would be hosted all on someone else's server with self hosting diminishing the ease of collaboration.
Hm, I don't think online tools are that far ahead. If I look at the horrible experience, which is Google docs or MS Office 365, especially their word processing parts, they lack behind soooo much compared to a tool like Libreoffice. You cannot even properly use styles and, if I got a few $currency for every silly bug I encountered in those online half-assed tools, I would probably work half time only. If that experience is anything to go by for the design space, then I would rather work offline and without collaboration, than having to worry about my whole document getting completely messed up by some silly bug.
> If I look at the horrible experience, which is Google docs or MS Office 365, especially their word processing parts, they lack behind soooo much compared to a tool like Libreoffice. You cannot even properly use styles…
As a reasonably serious user of Office 365 and someone who's desperately wanted to like LibreOffice, I'm not sure what you could mean by "properly use styles". I'm guessing this is specifically a critique of the Word web app, but the native Office apps also do "online, live collaboration", and it all generally just works. Even Apple's overlooked suite feels more mature (and offers a dramatically better user experience) than LibreOffice.
Depends on what you want to do... I find LibreOffice fine because if I want to create documents that are actually complex I'm going to reach for InDesign or LaTeX. I don't find something like Microsoft Word really much better for anything I'd want to do with it than LibreOffice.
If I just want to make notes, I'll use Org or Markdown in plain text. These can also be converted to HTML for web.
If I want to print something with lots of text, nothing beats LaTeX.
If I am making something design intensive, I'll use Scribus or some other desktop publishing software.
I guess regular people think of word processors as the Swiss army knife here: does everything, not particularly well, but worth only having to learn one software?
Libre office is pretty much the only game in town on Linux. Well excepting the web based office suites. I really like it, especially for csv files which it works great for.
Hey now it’s unfair to limit this to just their online verions. I have reproducible word bugs that date back to word 2007.
Microsoft in my opinion is a company that writes good core logic followed by trainwreck UI/UX teams only eclipsed in their destructive cabability by the user hostile marketing and profit department who keep trying to get me to do stuff like upload sensitive documents to OneDrive through some absolutely intended bad UI, or default to the most expensive resource selection in Azure every time you hit the back button on their forms.
In contrast to your anecdote, I have never had issues with Google docs. It has always just worked, reasonably clean interface. I'm not sure what elaborate styling you need, but it seems serviceable for when you need a basic document which I imagine is most peoples use case. Also the collaboration is actually very useful. The same reason I use overleaf as an interface for a lot of my LaTeX projects. I can quickly share and get updates, they just have to open their browser instead of download pdf > open it in their pdf viewer > make comments > send it back etc. The web interface is great for people that do not use git etc.
That’s exactly the problem; Google Docs’ use case is only “share a basic document and collaborate on the words.”
It’s not capable of creating precise or maintainable documents (e.g. no styles as others have mentioned), which is a very common need that it’s adjacent enough to that lots of people use it for and then it fails.
What is a style? Are you trying to create a poster or do some graphic design? Some weird xml style sheet thing? Yes it’s not for that, but then also doing that in libre office or word also sounds horrible and unmaintainable.
To define a custom style in a word processing context means to define "a group of settings that I can apply at will to parts of the text".
That is, I should be able to define a "Code block" style that should appear in the same dropdown where google docs lets you select heading levels. When I select that "Code block" style, the app should format the selected bit of text according to the syte's definition. For instance: use a monospace typeface, apply some margins to this paragraph and paint a gray background on it.
Likewise, I should be able to update the style definition and all code blocks should update accordingly.
This is very basic functionality that has existed in word processors since... forever. Large and/or more complex documents are simply much much more painful to maintain/evolve without this feature.
To add on to your well laid out points, many companies have "house styles" embedded into their templates that allow a consistent look and feel. GDocs lack of support for these makes it a non-starter.
gsheets lets me collaborate with others in real-time and share with others very easily. When you think about “performance”, you have to consider more than just the immediate steps taken while in the application. If you think about the entire end-to-end experience, gdrive is so much faster and seamless.
office 365 offers "co-authoring" and easy sharing as well. You're talking about something excel already has.
But working on large files is ridiculously painful. It's just not as as good as a native app. I don't want to give up my ability to work on large files in a performant fashion, which I do all the time, in exchange for collaboration support I do occasionally but a worse overall experience and frustrating lag compared to a native application.
Excel formulas have a horrible 1970:ish syntax. Reminds me of BASIC programming. But typically I get the job done, sometimes after several awkward intermediate steps.
I wasn't aware that Google has a different one. I have never used more than simple sums and differences in Google sheets. Can you give an example where Google offers better, incompatible functionality?
For example, QUERY [1], SPLIT [2] and ARRAYFORMULA [3] are the ones that comes to mind. To be fair MS Excel can do arrayformulas, but I prefer GSheet's syntax. Last time I tried Excel, TRANSPOSE wasn't a thing it but it seems to be there now.
I periodically check on LibreOffice to see if it's usable and typically I don't last more than a minute or two without hitting a major show-stopper. Last time I tried LibreOffice Writer, the text didn't render until after I stopped typing. The time before that, LibreOffice Calc took 3 seconds to begin editing any cell.
It's a common problem with Open Source: who does the slog work of testing, tracking down, and fixing the simple bugs and perf issues? Evidently, LibreOffice doesn't have enough maintainer cycles to have good coverage of common setups. It's unfortunate and as a developer I feel for them, but as a user it makes paying Microsoft an easy choice.
The original comment was about the design space, and I have to agree: Inkscape is stuck behind. Figma is a very high quality product, actually fewer bugs or issues than its offline counterparts.
I really think that the future is Local-First software and FOSS can greatly benefit from the new paradigm [1]. With Local-First software the cloud is just the natural extension for backup and collaboration (real-time or offline) not the sole repository for the data.
[1]Local-first software You own your data, in spite of the cloud:
The concept looks fine, but was there resistance from users to have local apps talk to cloud resources in the first place ? It looks to me like it’s just describing the situation from a few years ago, but as a pre-packaged concept.
I understand the current situation as a result from evolving business models, that lead to better profits when the control is taken away from the user and they have to get into subscriptions for continuous use. Basically the Adobe CS story. This manifesto doesn’t touch the business aspect at all, so I’m not sure what to take from it.
Coincidentally, currently there are ongoing discussions on how to utilize Git outside of the software engineering field.
In particular, this is small anecdotal case study on how a calendar booking/rental system can work with Git [1]. For me this is a wonderful example of application for Local-First software albeit a simple one.
[1]Ask HN: Why hasn't Git been adopted outside of software engineering?
Once you have your centralized cloud service integrated, you have ongoing support costs so you must charge subscription instead of letting your users fully own the software. The only acceptable paradigm in my view is local-first with P2P collaboration options.
I don't agree that this thing about moving to the browser is currently relevant outside of the drawing software like figma and equivalents. I don't see NLE, 3D suites, DAWs and many more moving anytime soon to the browser. Maybe photo editors like photoshop? But even that isn't there yet.
And to be honest the only FOSS content creation software that actually competes at the professional level (i.e not just simple edits) with proprietary tools is Blender: As in, would people use this software, if all the (currently proprietary) alternatives were also FOSS? But inkscape doesn't need to compete. So I don't think it's getting left behind in the dust.
I'm sure too that most inkscape users would much rather have a lot other improvements, than moving to the web.
I would still use inkscape, it's way faster and lighter and though it crashes sometimes, it doesn't corrupt files like Illustrator. It also has (IMO) a simpler interface that's somewhat more discoverable than Illustrator but not by much.
Yep, exactly. It's absolutely insane we've now reached for a point where when people see a standalone, working, polished piece of software that respects your freedom, they somehow feel that it is "inferior" because it doesn't take away your rights.
I, for one, am glad that Free/Libre software like inkscape continues to be heading strong, and find that it far outpaces any of the proprietary malware SAAS applications out there.
It's not insane. People want to know whether free software package X is a good alternative to the proprietary version. If it isn't, the fact that it's "good enough on its merits" is irrelevant. If they don't think proprietary software is intrinsically evil, then the free/proprietary distinction is irrelevant to them too.
In what way do you find Inkscape to outpace e.g. Figma? That would be interesting to hear.
The free/proprietary distinction is only irrelevant for small-ish projects, on the order of a few months, where "good" is a one-dimensional measure in terms of features the application has: The risk of the software maintaining company being bankrupt in this timespan is low.
A typical mid-level employee at a company might be very happy with such tools, especially because they are usually new & shiny too.
Now imagine you are part of a team that wants to solve BIG problem, over a LONG timescale.
Suddenly the picture becomes more nuanced as software features aren't the only thing that matters anymore.
Would you really want to make the success of your project dependent on a third-party software company whose objectives aren't aligned with yours and that can take the tools you use away from you on a whim?
FOSS makes a very strong point there. The interface may not make you cry and melt your eyes with shiny buttons, but you can be sure that years down the road the software will still be there and you can read your old files no matter what, as the file formats are open. You can depend on the software, and,.more importantly, you can assess and proactively mitigate the failure modes in advance!
(E.g. the tex files from the 90s that you can find on some researchers website can still be compiled today; I imagine the entire research community would be significantly hindered if they would have to deal with the zoo of formats of proprietary software, rtf, doc, docx etc. In this sense FOSS can even act as the catalyst for establishing a standard, as is the case for tex files in domains of science.)
Though I agree that for image editing specifically, typically projects are probably not longer than you few months, so in this you case I imagine you could go without issues with the flow of new and proprietary software.
If FOSS is responsible for establishing TeX, then I would pay Bill Gates my last $ to embrace, extend and extinguish the entire miserable movement. TeX is a horror from the 1980s that refuses to die.
That's not exactly a FOSS problem, though. Nothing stops an open source project from offering the same collaborative features as proprietary solutions.
It just seems that people who work on FOSS value privacy, ownership and freedom more than braindead ease of use and monetization.
That seems to be an unfair dismissal of these products. The vast majority of everything built with these tools will be public facing. Marketing designs, UI design, etc.
Design is a very collaborative process, especially with remote work which has made having everyone see and edit the same thing at once critical. It's not "braindead" to work in a team, its a requirement. I really want to see FOSS succeed but its hard to see how the open source benefits translate over to this kind of software where you no longer run the software locally and consistency is needed so others have the same version of the tool as you.
Sandstorm.io tries to make collaborative web apps work easily with sharing for non-technical users, though we haven't solved getting people a server easily, or more crucially, convincing people to buy a server to do things they can do for free with ads.
Unfortunately, the problem is companies powered by ad revenue have a lot more money to throw at usability than FOSS folks do.
> we haven't solved getting people a server easily
It doesn't help that Sandstorm focuses on apps that are architected with a devops-heavy mindset. Many apps don't really need anything more complicated than PUT and DELETE (no POST special handling required), esp. since they're already doing so much on the client, anyway. This really calls for an auth-protected document store more than it does what Sandstorm tries to provide; the difficulties of "getting people a server" probably starts with assuming/insisting that that's what you need to do in the first place.
Those kinds of architectures are only necessary when you really do need some way to offload computation to a remote, high-availability machine, but things like online image and document editors aren't that. (Even then, what you really want for that is some sort of mobile code architecture that lets you squirt lightweight payloads at some sort of fleet of faceless workers, preferably in the same/similar form to the kind of code that you write for doing useful work on the client...)
I really don't agree. I think loading a bunch of client-side crud from a random web server and storing it in opaque browser buckets is super untrustworthy. I think to some degree modern web app development has gone off the rails a bit.
Why are you moving the goalposts (ignoring the premise)? Sandstorm apps (like e.g. Etherpad, advertised right on the front page) aren't already doing lots of client-side work (locally), like I originally said?
All of these apps and products I mentioned do not have ads, they are paid for directly. Users/businesses do want to pay for software, but when they do pay, they want it to just work seamlessly. No one wants to set up and maintain a server.
I mean, ultimately the answer there is... pay someone to run your open source server for you. But since the OSI hates SSPL and the like, the actual open source developers are disadvantaged in running such a service compared to someone who just downloads it and hosts it.
A huge issue is that the gatekeepers of "open source" don't believe making it financially viable is something they need to address. So it's gonna end up being the best we can do in our spare time for the love of progress.
Do you know Penpot? https://penpot.app
Its like, sketch/figma online, collaborative and open source. Its not at the same level of Figma but its pretty good!
Users are given a choice. There are mountains of alternatives to adobe products. But they keep paying for the adobe one because it’s the best. And it’s the best because people keep paying their monthly subscriptions to fund it.
They might complain but at the end of the day they are very happy to pay adobe for the best in class tools.
The industry keeps paying because of the sunk costs of legacy files and content, and the fact that Adobe also makes RIPs for imagesetters so the expectation is that it will all be compatible and work together.
As a proof of concept about the possibility of a design tool with the potential to share information in this way it's intriguing.
I'll also add that a major pain point of FL/OSS tools for me is that so many of them HEAVILY depend on local file descriptors which makes remote storage access a PITA.
> The design space has been very rapidly moving towards online
I think the opposite will eventually be true, a browser is no space for complex documents.
> Even a lot of the big name proprietary tools are finding themselves becoming rapidly obsolete in a world where people want to IM a link and just have it reflect the latest version rather than attaching files back and forth.
O365 has worked like this for years and it works well.
> The design space has been very rapidly moving towards online, live collaboration tools like Figma, Sketch, Adobe XD, etc and FOSS just has no answer for this market.
Some people speculate whether the rise of SaaS is a response to FOSS, which made it difficult to monetise software development. I’ve certainly begun to question my previous automatic assumption that FOSS is a good thing in the long term.
The good news for FOSS is that hosting costs just keep getting cheaper, so providing free SaaS is increasingly feasible.
There’s also the “third way” of FOSS with a hosted option available. Wordpress is a good example of that. If deploying such projects on the cloud eventually becomes trivial, then you could imagine a FOSS desktop in the future might have a cloud component (or be entirely remotely hosted). That might sound odd, but consider you still need to buy hardware to run your FOSS on, so renting that hardware instead isn’t a great leap.
Wordpress is an example of that for the simple reason that it is a CMS, which by its very nature implies it works in a server-client fashion. Of course Wordpress can be run as SaaS or self-hosted–it is software that needs to be hosted to serve its purpose.
On the other hand, editing vector graphics has no intrinsic need to be hosted. Just because a cloud-based FOSS desktop in the future could be imagined doesn't make it a reasonable idea. What would be the point, other than massively increasing complexity and points of failure in terms of reliability and security?
Lots of points, but the most obvious one is with SaaS you can deploy a single version of software, which makes it orders off magnitude less difficult to maintain.
I don't understand that argument. I sometimes use Inkscape but I never have more than a single version deployed on my device. Also, I don't think I'm putting any pressure on the maintainers if I were to insist on never updating from an older version.
Snarkiness aside, being able to "deploy a single version of software" is itself incompatible with the ability to self-host, making it a moot point for FOSS (let alone for FOSS that has no practical reason to be SaaS in the first place).
Until you have users that want a feature in an older version that's not in a newer one. Oh, and companies are always saying "nobody uses feature X, it was hard to maintain so we got rid of it" which is infuriating if you're a user of feature X.
Good employees find ways to collaborate no matter what. That’s to say that real-time collaboration is a luxury feature. Also, many projects have only one designer. Not to mention, it’s rare to find designers that work in tandem.
Also, from a dev perspective having designs updates without some process is undesirable.
As a different point of view, I work w/ my designer in Figma, and the collab features are a must-have. Our iterations are much faster in Figma than they would be on static files shipped via PDF, or whatever it would be.
I've worked with designers in tandem as well, sharing screen, sitting next to each other, and in online real-time. Real-time collaboration is a way to collaborate, not the only way.
Knowing how to collaborate well with LaTeX makes you about as "good" an employee as knowing how to collaborate well with Inkscape.
The point is that even in places where about half the employees know how to collaborate with these tools, it is not done because you need the other employees to function.
I meant good as in good faith, but I see your point. I started using Corel Draw ~97, and after trying many design programs of all them seem about the same complexity-wise.
Now, many people tried Gimp (which is far from Photoshop) and they wrongly think that Inkscape is another Gimp.
I have never used photoshop and I still feel unsatisfied with GIMP. I have a long list of thoughts but the simplest one was the fact that moving/rotating/scaling a layer is 3 different tools. I was told a long time ago that there was work to merge them, years later the change comes out and it just groups the tools under one icon so you now have to right click first to select the tool you want. It's pretty much broken on a 4k display as well.
I do like gimp as a bit of a Swiss army knife program, it can do a lot of utility jobs like resizing, outlining, etc which can be hard in some of the more intuitive modern apps, but most of my work now is done on an app like Procreate and then I'll move to gimp for the final edits.
Gimp certainly has its GUI problems - coming back to it after using Darktable has been quite painful. I can accept it has GUI issues. However, for what it does, the only tool that is better is Photoshop. Sure, Lightroom/Darktable has a lot of nifty convenient stuff with a better GUI, but they're simply not a viable replacement for Gimp.
There are other photo editing tools with better UI, but poorer features, sadly. It would take a ton of work to catch up in terms of features.
> I have a long list of thoughts but the simplest one was the fact that moving/rotating/scaling a layer is 3 different tools.
You might find that fact to be on the ageing side :) Unified Transform tool has been available for many years now. It's all three (and more) tools in one.
If anything Gimp is too much like Photoshop. Corel Photo-paint has a much more powerful paradigm of manipulable bitmap objects rather than shoehorning everything into fixed dimension layers. It's unfortunate it never got any respect and withered.
The point is that few workplaces have a majority of employees that are "good enough" to find ways to collaborate with Inkscape. If even a third of them cannot find a way to do so, you're stuck, and siloing your/their work is not an option. No decent manager would accept that merely because you insist on using Inkscape.
Sure, but once you make the effort required to build a real time app, there are vanishingly few reasons why not to have it as a web app. Some of them provide a desktop app as well but they all provide a web app.
As a designer, you can just send a link to your design to anyone, end users, clients, developers, managers. And they just click it and it shows up and works. No need for installing software, setting up accounts, etc. A lot of these web apps are actually C++ and then compiled to web assembly so they are quite fast.
Do you use source control or do you email zips with the source files between you and the other developers? The collaboration here is exactly the same as that distinction in workflows.
It's obviously not just "two clicks", it's "Text copy 2 final final alt reviewed.doc" and then you get a new person email their suggested updates to you as "Text copy 2 final 2.doc" and you have to somehow merge those suggestions in with the head version.
With Word at least there's the ability to do suggestions rather than direct edits to let you collaborate against nonsynced docs at all, but with Inkscape when that happens you'll literally just get a different svg and you can't actually merge at all, instead you'd need to conceptually understand what edits were made and then make them again manually to the "head" version, which is completely redundant work.
This post is about Inkscape and Inkscape svgs dont have any edit history baked into them. If you had a dozen revisions and one of several collaborators sends you a suggestion version which they edited from some unknown revision what would you do with that? The file itself doesn't imply what deltas they made, and "copy paste" doesn't make sense, since what would you even be copy+pasting?
Yes, I honestly think that if you have collaborators doing edits forming a DAG of derived artifacts, and you only observe a subset of those artifacts that you can't take some arbitrary artifact and merge that correctly with another arbitrary that opaquely branched some point in the past. And that is what consistently happens with the "just email some zips of the documents" collaboration flows.
Git helps by forcing the deltas to be knowable, but at the expense of a complex management tool and only really working on text based source. All other modern software now solves this problem by being online collaborative which eliminates the problem entirely, and that benefit dwarfs any other feature for most corporate/organizational usecases.
Inkscape is great software, but it's not acceptable software if you're going to collaborate with multiple other people who will all be making edits to the same document.
Virtually all FOSS projects grind to a halt with regards to basic features as soon as they gain traction. The old userbase wants to keep the UX, others want to revamp, yet others want some other kind of pivot.
The real pivot window is before you gain traction, unfortunately.
This seems accurate. Reminds me of libre office which has a settings page to pick between 5 different toolbar UIs. I have got to the point where I don’t want to be customising my toolbar. I just want the default one and I want the default one to be the best UI. While the default on LO hasn’t progressed in 20 years.
Do you ever think a project like Blender will end up like this. I don't think there is anything remotely similar to it online and I doubt there ever will be.
I think it will when the web becomes powerful and efficient enough to run such a tool in the browser. I don't think its there yet but I'd expect it will be eventually.
I'm not talking about low level infrastructure, but the user facing software. Offline desktop tools have rapidly started to become obsolete. Products like Office had to scramble to pivot because Google Docs was replacing them. Its been a while now since I have been emailed an actual office file instead of being sent a link to the live version.
A while ago it was common for Photoshop or illustrator to be used for UI designs and then it would be thrown over the wall to you as a PNG which you had to make sure you had the latest version of. These days everywhere I have worked has used a tool like Adobe XD, Figma, Sketch, and gives out a link which is always up to date, which you can fiddle with, write comments on, etc.
The market is rapidly shifting and I'm just not sure how FOSS is going to stay relevant in this space.
> The market is rapidly shifting and I'm just not sure how FOSS is going to stay relevant in this space.
How realistic is it to expect FOSS tools to directly compete head-to-head with these incredibly well-funded privately owned tools? The whole point of these companies is to solve the "hard" problems in order to gain a functional and competitive advantage in the market. FOSS OTOH deals with table scraps in terms of funding and technical talent. And let's face it, who are some of the biggest contributors to open source? These very same companies. They're happy to help keep the specific libs that matter to them up to date, but they're certainly not going to give away the entire product for free.
> Offline desktop tools have rapidly started to become obsolete. Products like Office had to scramble to pivot because Google Docs was replacing them. Its been a while now since I have been emailed an actual office file instead of being sent a link to the live version.
Just because the market moves in a particular direction doesn't mean it's the right decision for end users, all it means is that it's the right decision for the business. Take the mobile phone manufacturers removing the headphone jack. No customers asked for that, it was purely a move to drive sales of bluetooth peripherals.
There may come a time when running software locally becomes in-vogue again, who knows? It's companies that get to dictate the platforms because they own them. And as others have pointed out, it's not like these online collab tools aren't without their own issues. I guess I just don't understand why people think that FOSS can deliver polished end-user products on the scale that companies do.
Why can't FOSS do this, though? There's no reason Inkscape couldn't work the same way, where a document lives in the cloud and users pass around links to it.
It does require someone to maintain and pay for the cloud hosting. That shouldn't be unsolvable; that's not really any fundamentally different than hosting a download site for traditional FOSS. Donation-supported FOSS can work.
1. A generic CRDT library, preferably in C or something that exposes a C interface. This is probably very hard or even impossible, depending on how much must be abstracted and how much work the developer must do to express edits in terms of the CRDT.
2. A P2P backend for said library for streaming edits.
3. A networked data store, maybe as-a-service, for versioning and persistence.
2 and 3 might even be made transparent to non-CRDT-aware applications through FUSE, but that's not a hard requirement.
If the software you use is decently priced and does what you need then fine. I'm sure that support is fine too.
When I call for help in say Krita or Samba on the mailing lists, I often get the actual programmers responding. Not all the time but quite often.
I've been running an IT company for 22 years. I have a fair idea about what the market is up to. It turns out that a wanky LDAP n Kerberos combo will still be available next year, as it has for the last ... (when did Windows NT domains morph into AD? What a bloody nonsense)
It's not even clear how FOSS would even work in this market or how it would benefit the user when it would be hosted all on someone else's server with self hosting diminishing the ease of collaboration.