I really like the author’s listing of advantages open source provides that you just don’t get with closed source, it aligns very closely with my point of view. This is great reference material for my little archive of “read this to understand my point of view”.
I’m also skeptical to what GitHub’s social coding concept has done to maintainership in general. There are projects that don’t need frequent updates, and projects that benefit from being the vision of a single person. Critiquing maintainers for refusing to let others in puts unnecessary pressure on people who just want to share something they’ve made with others.
Reminds me of Linus' comments[1][2] on pull requests in the kernel's GitHub mirror. The fact that you can't disable pull requests give the impression that it's mandatory to allow anyone to request changes and for you to review them.
Yeah, especially as you can disable all the other tabs (beyond code). I think that’s been successful in encouraging open contribution projects, but that has also impacted the notion of what open source means to people in a way that isn’t just good.
Oh that looks like a nice product ! And the author’s change in reasoning is interesting to read.
At the end of the day, it’s just the internet. It’s full of cool people but also full of toxic people and you have the right to just ignore them.
Working on an open source contribution without discussing about it beforehand and being angry at it not being merged always have been a rude and toxic behavior. And like "irl", you have the right, if not the duty for your own mental health, to ignore toxic people.
And I don’t mean that it’s rude to create a pull request without prior discussion, it’s ok and sometimes faster for everyone, but when you do this, you have to remember that maintainer(s) doesn’t owe you a merge and that if it’s just closed with an explanation, it’s life, even if you don’t agree with the explanation. Also if your PR was 10 days of work, that’s your own problem.
Another thing you can do is to publish source code for releases only.
By keeping the commits internal you don't have to think hard about what you commit and it allows to have other stuff in your repository that logically belongs there but is not suitable for a public release (internal tooling, things used to create the source code initially but not required, etc.).
I am a Bruno paid user, and I appreciate its design philosophy, such as being local-first and using a text-based DSL to represent request structures. However, the experience with Bruno has been somewhat unsatisfying. It takes a long time to start, and the UI always feels laggy, not to mention that it significantly increases the temperature of my M1 Mac.
I tried Yaak today and found that it not only has a smaller package size (160MB compared to Bruno's 300MB) but also runs very smoothly. The UI is so intuitive that it initially seems like it might be missing some features, but I soon found that it has all the necessary features to support regular use cases. I hope Yaak can support Git sync, and if it does, I will replace Bruno with it. Nevertheless, I will support Yaak if you have any business plans; it is a beautiful app worth paying for.
I think those who care about things being open-source care deeply about it and it is an issue that will make them avoid something. However, that is a very small group of people even in the tech world it's pretty small.
Businesses don't care if you're source-available or open-source and decision-makers often won't know the difference. Generally, people looking for open-source are just looking for something free they can use/self-host.
I agree with this fully. There is a small sweet-spot of people who are doing hobby projects or have SMB tech companies to where a project being open source matters. Note: this is much different from infra being open (React, Linux, etc).
People talk about lock in but it's way cheaper to migrate to a different solution than to maintain an entire product of it becomes unmaintained (i.e. an "open source Stripe" where you have to keep up with compliance)
I understand that the visibility causes better security but there is no guarantee that the hosted version is using that code.
Many billion-dollar companies barely have in-house web developers, let alone someone who can set up a K8s instance or a Hetzner server. "Little tech" is just such small portion of the economy.
As a decision-maker I care if something is open source or not. Because if it is, I can always fork it and maintain my own version if absolutely necessary. Not something I ever want us having to do but knowing the option is there is huge. I don't want to be beholden to another corporation that could pull the rug out from under me. Open source software gives me that comfort.
I did say "often won't know the difference" and didn't make an absolutist statement. And your comment basic confirms what I said, those who care care deeply and it's an issue they'll walk away from. But as I said, it's a very small niche.
Kudos, congrats and thanks for open sourcing your stuff.
Your openness to feedback, even if not quite pleasurable, is impressive.
I believe no contribution is a good tradeoff that still guarantees user rights important to the free software movement. It is important that developers don't conflate free software and open to contribution. If managing contributions is not fun to someone, that shouldn't prevent them from releasing under an open source license, or force them into managing contributions.
And if someone releases a version that you like, you can always upstream this stuff yourself if you don't require a CLA or anything like this.
I will look into this thing, it seems useful.
> Now the desktop Linux users can fix their own bugs (please don't hate me)
I've not heard of this before and upon looking this looks fantastic! Cross Platform and sticking to being good at requests!
Postman really grinds on me as to how it's gone down a path of forcing "collaboration" when sometimes I just want to ping some requests and test things!
Great question! (Creator here)
Yaak is meant to be just an HTTP client, where Postman aims to do everything related to building APIs (mocking, testing, etc).
This makes Yaak simple and easy to use, in comparison
I'm not sure where the derision is coming from. It seems like they made an assumption about something they have experience with, got a bunch of feedback, changed their behavior, then took the time to write about their journey for the rest of us.
The author is engaging with everyone in good faith. This kind of openness and willingness to change would be applauded.
The author could have simply written nothing, convinced themselves they were right about staying closed source or...
The author could have just ignored all the feedback they got and stayed closed source or...
In the face of opposition, really dug in an tried to double down and further justify staying closed source or...
Take the time to sort the constructive criticism out from the negative garbage, digest it, and try to grow from it.
They should be commended for being open minded and putting in the effort to grow. And to directly address the GP's comment, in order to grow the conversation has to start somewhere. You learn/grow through experience/engagement, and that's exactly what the author is doing, and in a pleasant, tactful way.
As I said, this is the Internet, so there is certainly nothing wrong with it, and there is an audience for these kinds of posts. It is just not me, and the comment is probably better understood as a remark to myself.
For some context, this has been posted in reaction (author's admission here [1]) to yesterday's reddit thread "Postman is shit - non-enshittificated (OSS?) alternative?" [2].
In that thread Bruno [3] is also mentioned as a non-curl replacement.
I've been using Bruno for a couple weeks on a project, and one thing that I've been enjoying is that I can save all the request configuration in files right next to the source, so they are part of the repository. I tried looking to see if you could do that with Yaak but just looking through the documentation, I didn't see any mention of it.
I still wonder why people do this. Build an end-user product and give it away for free.
It’s just making our work look worthless and kills small software companies.
Cool, share your library or web server, as it allows others to build products around this and you will benefit from theirs. But consumer products? Why are we so cheap? Will we do just anything for „likes” and „internet points”?
> I still wonder why people do this. Build an end-user product and give it away for free.
If you don't intend to monetize it at all (or don't think you can), there are other potential benefits:
◌ People using it can draw their attention to your other output.
◌ People using it can be a free testing resource, finding bugs before they affect your own use of the software. They can also be a time sink due to support requests, so this can be a delicate balance.
◌ People using it can be a source of improvement ideas that you have not thought of but which are useful to you once identified.
> It’s just making our work look worthless
Only if the sole way you judge worth is financial income.
> and kills small software companies.
Sometimes this is a deliberate part of the equation, when a large company opens some of their product(s), but usually not.
If your company is surviving on selling a relatively small end-user product like this then you are in danger from other small companies doing the same anyway, even ignoring any F/OSS options, and need to be competing at least in part on other factors like support, quality, and keeping up with changes in the industry (or your users' industry if that differs) so your product stays relevant and useful.
Open source and gratis doesn't make our work look worthless.
First, money ≠ value.
And even when talking about money, look how are valued many open source projects like MariaDB (in the news recently), nginx, WordPress, … - yes, caveat, you mentioned consumer products, those are not all consumer products)
For some of us, it's not about likes or internet points. It's not being cheap. It's about giving the users the rights they deserve. It can be about efficient cooperation (here, it's not about this). It's also becoming a requirement in some contexts and places.
There are business models around open source software. There are other ways to make money with software than to sell proprietary software licenses.
Of course money != value, but don't pretend that REST client is charity or fight for freedom or something. It's a wrench. And wrench should have its price on the shelf.
Unfortunately most business models around free software are ads and user profiling. Rest is just noise.
The wrench analogy doesn't work because when downloading a copy of the software, you don't make the original vanish. A copy costs nothing. There are also costs associated to software maintenance that don't apply to the wrench. Software has its own characteristics.
But I think I get the essence of your remark, and that would be that work needs to be paid. I mostly agree (I agree "enough" to carry on anyway), more on that later.
> Unfortunately most business models around free software are ads and user profiling. Rest is just noise.
That's unfortunate but not an inevitability. You don't need to rely on ads for revenue. It's not because many businesses decided that it's okay to spy on their users and steal their time and attention that you are forced to do the same.
The projects I mentioned don't make money through ads. Other big projects are like that as well. WordPress and Nextcloud for instance both have different business models that don't rely on ads. I work for XWiki SAS, we also don't rely on ads. There are ways. You can even sell "licenses" for open source software. We do that (and it's not open core), although I think you complained more about the gratis aspect than the open source aspect. We also make money with consulting, support and hosting.
> don't pretend that REST client is charity or fight for freedom or something
I don't see what you mean. If I'm to rely on some tool, I want to be able to control it and proprietary software doesn't allow me this. So I won't use a rest client if it's proprietary. And I wish this also for my users.
Yeah, maybe I've got too old-school thinking. Or maybe just worked on too many failed startups :). But the overall cheapness of our industry is driving me crazy.
I happily use proprietary software - IntelliJ, Sublime, Affinity, MacOS, etc. In my experience if you get software for free, you are the product, or you will be charged x10 more when they finally get their "business model".
- if it's free software you use locally (or SaaS you know you can self host), you should be able to keep using it even if the business model changes (which can happen in the proprietary world, see Adobe for a spectacular example). If the software is popular enough, there's some chance the thing will be forked and maintained by somebody else. Of course, no guarantees on this. I strongly prefer local software for this reason.
- you should definitely look at the business model of whoever builds your software. If it burns investor money or runs on ads, it's not good, for the reasons you mentioned.
- paid software is not a guarantee against your concerns. I already mentioned Adobe, windows is also full of ads and spying nowadays. VMWare and Confluence also have fucked their customers quite hard on pricing recently. Actually, I quite strongly distrust proprietary software seeing how common these things seem to happen in the proprietary world and there's no "someone is going to fork it" backup plan.
MacOS isn't without issues neither. It forces you to login for software upgrades and to access the app store (that's tracking in my book), and it's game over when Apple stops supporting your hardware. Its proprietary nature pretty much guarantees nobody will take over. Perfectly good hardware becomes insecure or waste unless its owner decides to replace the OS with Linux. Which isn't possible on iPads and iPhone.
On my end, I haven't had bad surprises so far. Ubuntu has been a bit worrying but I've been using something else for years now. The biggest concern for me today is Firefox. The strictly open source nature of the repositories from which I install Firefox is a first "safeguard".
My software is largely sponsored by companies selling support, they have an incentive to make sure it keeps working. It is so good that I can install the latests versions on 15-20 year old hardware.
Yeah, exactly because of that we have no nice tools for devs. All is built either to get a better job (and then abandoned) or you cannot sell it (and then you abandon it). If we were not that cheap maybe we would have more Jetbrains like companies.
Every company is replaceable.
I just don’t see doctors or lawyers giving away their work for free. They are just smarter than us.
I just don’t see doctors or lawyers giving away their work for free. They are just smarter than us.
"For free" does a lot of heavy lifting here. Of the members of each field, only a fraction of these professionals are doing primary discovery & research in their area of expertise. The rest of them are leveraging the completed research/studies/trials/precedents of others from their fields, in some topics decades worth of it.
This is the first thing I looked for on the website, FYI - any indication of what the long term plan for Yaak is. It might be worthwhile to put a note of "cloud sync coming soon" or something, or maybe I'm an outlier and it doesn't matter.
I’m also skeptical to what GitHub’s social coding concept has done to maintainership in general. There are projects that don’t need frequent updates, and projects that benefit from being the vision of a single person. Critiquing maintainers for refusing to let others in puts unnecessary pressure on people who just want to share something they’ve made with others.