 | Logseq: Feedback | 4173 Members |
| Discuss improvements to Logseq here. Want to propose a feature to the team? Please use the forum as this chat is not actively monitored by the Logseq team. https://discuss.logseq.com/c/feature-requests/7 | This is the feedback channel of the official Logseq Space on Matrix, which you can find at #logseq:matrix.org. | 18 Servers |
| Sender | Message | Time |
|---|---|---|
| 26 Apr 2024 | ||
 android876 | In reply to @_discord_705382494819844168:t2bot.ioOh this is great I was about to ask if Logseq works with Syncthing because I have used it and have had success but I cant help but feel insecure about if Im setting it up right and if theres a guaranteed way to set up Versioning setting thing | 02:28:44 |
| android876 | Personally I would have avoided using commercial cloud services from the beginning because firstly most companies dont have the most basic practices such as storing peoples passwords in Hash form. Secondly the cloud services of today are mostly closed source and non free (As in anti freedom) anyway and I think with todays tech we should avoid all blind faith in software (closed source) and use only FOSS since its easier to have large groups of people collaborating on te source code to fix vulnerabilities and such. | 02:34:16 |
| android876 | Sycthing has been working pretty well for me but I wanted to ask if anyone else is using it without any failures | 02:35:35 |
 crazyps. joined the room. | 04:16:30 | |
 alxlg | feel free to fix the issue—Logseq is open source after all But Logseq has a CLA: everything contribution to the codebase could in the future be turned into a proprietary product and it is happening a lot recently. See Tdraw and Terraform. Tldraw in particular affected Logseq too, since Whiteboards is now stuck to an old unmaintained version of Tldraw. Many open source contributors realized the danger of CLAs is real and won't contribute anymore if that requires signing a CLA. | 09:05:23 |
| alxlg | * feel free to fix the issue—Logseq is open source after all But Logseq has a CLA: every contribution to the codebase could in the future be turned into a proprietary product and it is happening a lot recently. See Tdraw and Terraform. Tldraw in particular affected Logseq too, since Whiteboards is now stuck to an old unmaintained version of Tldraw. Many open source contributors realized the danger of CLAs is real and won't contribute anymore if that requires signing a CLA. | 09:05:41 |
 adriand_35201 joined the room. | 11:57:52 | |
 thisanneke joined the room. | 12:27:02 | |
 Ramses | Personally I don't have much experience with SyncThing as I use Logseq Sync. But there's lots of posts about it here and on the forum (https://discuss.logseq.com/search?q=syncthing) | 13:04:26 |
| Ramses | You're much more versed in licenses than me. But when the license was AGPL, how many community contributions were there? It might be interesting for us to see that. But regardless, both here in the channel, on the forum, Reddit, etc. there's been valid concerns voiced by users and contributors (whether it be financial, via code, or content). I've brought up (un)transparency several times internally, but it's very difficult to tell developers how to work when I'm not a dev myself. | 13:09:22 |
| alxlg | The license IS AGPL... a CLA or comtributor license agreement is a document that everyone contributing code to Logseq's GitHub repo has to sign | 13:11:06 |
| alxlg | * The license IS AGPL... a CLA or contributor license agreement is a document that everyone contributing code to Logseq's GitHub repo has to sign | 13:12:27 |
| Ramses | Yes you're right. Again, I'm still out of my depth when it comes to licenses. It's not something I decide, that is with Tienson and the core devs. All I can do is gather feedback and try to get the devs' attention for it. But there's a lot of stuff for them to focus on unfortunately. | 13:13:28 |
| Ramses | That's also one reason why I think so much dev work happens internally. From a community perspective it would be much better if all development and discussions happened in public. That I've pointed out many many times internally, but there's no motivation among the team members to change their dev practices. I think they believe development goes much faster in a small team than involving the community. If this is wise, I don't know... I don't have that kind of project management experience. What I do know is that this makes it much more difficult to see where we are in terms of development and what users can expect, which I do not like. | 13:16:36 |
| Ramses | * That's also one reason why I think so much dev work happens internally. From a community perspective it would be much better if all development and discussions happened in public. That I've pointed out many many times internally, but there's no motivation among the team members to change their dev practices. I think they believe development goes much faster in a small team than involving the community. If this is wise, I don't know... I don't have that kind of project management experience. What I do know is that this makes it much more difficult for non-team members to see where we are in terms of development and what users can expect, which I do not like. | 13:16:59 |
| alxlg | Let me explain: copyleft licenses like GPL, LGPL and AGPL were designed by Free Software Foundation a delicate system of weights and counterweights from a political point of view. It worked great and made projects like the Linux kernel success. The issues start by upsetting the balance, mainly by using permissive licenses like MIT or introducing a CLA. With the latter, you give away the rights on your contributions to a company, that later can turn your code into a proprietary product: you don't need to be a dev to understand that this can break the virtuous collaboration mechanism. So to make things clear: the community-driven model that happened to be successful for the last decades has been the one using copyleft licenses and no CLA. I am saying this because other approaches are damaging the reputation of FOSS in general. You are free to use whatever approach you want, but you shouldn't invite people to contribute bug fixes "since Logseq is open source" when you didn't adopt the community-driven approach in the first place. | 13:29:07 |
| alxlg | * Let me explain: copyleft licenses like GPL, LGPL and AGPL were designed by Free Software Foundation as a delicate system of weights and counterweights from a political point of view. It worked great and made projects like the Linux kernel success. The issues start by upsetting the balance, mainly by using permissive licenses like MIT or introducing a CLA. With the latter, you give away the rights on your contributions to a company, that later can turn your code into a proprietary product: you don't need to be a dev to understand that this can break the virtuous collaboration mechanism. So to make things clear: the community-driven model that happened to be successful for the last decades has been the one using copyleft licenses and no CLA. I am saying this because other approaches are damaging the reputation of FOSS in general. You are free to use whatever approach you want, but you shouldn't invite people to contribute bug fixes "since Logseq is open source" when you didn't adopt the community-driven approach in the first place. | 13:29:28 |
 aaronbieber joined the room. | 13:34:05 | |
| Ramses | This is a great explanation, thank you. Yes, I fully see the problem now. Unfortunately, I don't see the team move away from a CLA as it could impede financial growth. Logseq is not a non-profit, and that's indeed where it could clash with the free in FOSS. So in that case I will no longer say to "contribute to Logseq is you don't like X". Thank you for helping alleviate my ignorance 🙂 | 13:35:26 |
| Ramses | Which makes the critical feedback from long-time users even more important, I think. Because not everybody is plugged into the community and/or GitHub and is wondering where things are moving. | 13:36:22 |
| Ramses | * Which makes the critical feedback from long-time users even more important, I think. Because not everybody is plugged into the community and/or GitHub and is wondering where things are moving and if bugs are going to get addressed or not. | 13:36:38 |
| alxlg | You are welcome! One last thing: despite the common idea of companies using permissive licenses like MIT or requiring CLAs, there are great examples of companies doing great with copyleft licenses and no CLA, like Red Hat, that is the one that made Linux so popular on servers. On the other hand, Canonical, the company behind Ubuntu, that uses CLAs, failed to make a lot of its projects adopted, to the point of later abandoning them in favour of Red Hat-sponsored community-driven projects (for the curious: Systemd vs Upstart, Wayland vs Mir, Flatpak vs Snap). | 13:46:31 |
| Ramses | Or Oracle? 😉 Because that's where I got my corporate scars. I'm somewhat familiar with Red Hat's case, having encountered them in my customer base (and often Oracle getting removed from a stack in favor of the RH products). Isn't their main earning model enterprise support? That's what I've pitched to Tienson several times, that we should focus our services on support rather than selling SaaS. That way, large enterprises could run their own Logseq Sync service and pay an annual fee to get support from support engineers on our side. But I haven't gotten much response to that so far. | 13:54:57 |
| Ramses | Though, it could be that he doesn't want to think about that until the DB version is ready along with improved sync and real-time collaboration. Because that's when we would be interesting for companies, especially if they can host the Sync/RTC service themselves and get support from us. That's just guessing from my side, but it would make sense as the guy has a lot on his head. | 14:07:59 |
 blogbourri | https://uxplanet.org/how-notion-grows-5cf152e81e46 perhaps a few more parallels in here than i expected to find. A key issue was that they had built their app on a suboptimal tech stack, and it was crashing constantly. Their seed cash was burning with no revenue to replenish it, and they arrived at a hard choice many founders find themselves facing — start over, or run out of cash. “If you looked at the burn rate, we all would’ve died together,” Ivan says. “It wasn’t much of a choice.” interesting article all around | 16:42:44 |
 b.l.t | more examples in this video https://www.youtube.com/watch?v=hNcBk6cwim8 | 16:58:50 |
 xarielml | I think it’s reasonable to expect devs to want to focus on dev work, not communicating that work. It almost seems like there’s a missing person in this core group; someone who can speak dev, understand the big picture, and knows how and when that should be communicated. Ideally they would manage the website to have these communications for broad reach (discord & the forum are great, but not a replacement for broad-reaching audience). | 18:57:45 |
 kiop2507 | I apologize for making you feel hostile, that is not what I meant. I don't have the ability to fix the problems with this open source software on my own, I'm just a very small white user, and I'm giving my opinion because this synchronization problem has been backlogged for a long time and has never been fixed. Obsidian of the same type has no problem syncing under iCloud. I'll tell you a little bit about my history with Logseq, I was impressed to use it after watching so many videos from the founder of YouTube. Think its a great concept. It has some innovative designs that I think are great too. But the biggest problem I've had with it from the moment I started using it to today is that I find it unstable. Often lost data, so that led to having to go back frequently to check for loss. But it's very hard for a human being to do that with human power. I've given you feedback on these issues, and I've emailed back issues before this. But finding this has been kind of a serious problem that has never been solved. You can certainly say that you're working hard to fix it, I'm simply expressing the true voice of the user. I'm not being hostile, but I do troll as many users do. I don't think you can just say if you think there is something else that works better, you can go ahead and use something else. Don't you carry emotions yourself when you say things like that? This is not a good way to treat users. When a product is born, it is bound to face all kinds of feedback from users, and you can very commonly think that I am one of those kinds of feedback. I'm not using obsidian as a complete replacement for Logseq, again, by my considerations as a user, obsidian is individually treating each day as a separate page, and this conflicts with my needs. | 19:03:36 |
| kiop2507 | This review is very pertinent. I can make sure I'm not that kind of hostile user, but you can put yourself in the shoes of a user who would troll that must have encountered a very large number of bad experiences in using it behind the scenes. I think the destiny of developers is to make the product better and better and more stable. I was initially impressed precisely because the founders had positioned the software very highly and molded it very well. I migrated my data to logseq because of this, but I'm not one of those high end users, and the way you responded to me is actually a bit of a high bar for a user like me. I think this kind of objective condition is also something you need to consider. I apologize for the misunderstanding of perceived hostility, but I hope you really feel my heartfelt trolling and that the emotions are 100% real. I used to drop a lot of data here as well. I've also recommended a number of friends to use it and they've dropped data. The other issue I'd like to talk about is that the content of the synchronized page is empty, but it's clear that the tab has been synchronized. This is a problem that I don't think should be borne by icloud, and I don't think it's something that users who don't pay for it have to bear. I think that's the problem with technology.☕ Ramses | 19:11:57 |
| kiop2507 | My English is not very good, so I am using translation software to keep carefully reading the developer's replies after the comments I sent. I also saw the message you sent. 🍺I think your presentation makes me feel solid and on point. My Chinese to English translation may be partly misleading, but I may not be able to check it. But I understand your English into Chinese, and I can probably understand it. I think what you said is very good. I think the way it is phrased is facing the problem head on. Emotions are not really that important in front of a product, whether it's the user's emotions or the development team's emotions. Emotions really don't even matter compared to the user experience. The fact that the problem can be solved should be the most important thing. Everything else is secondary. Otherwise the same problems that free users encounter will be encountered by paid users as well. | 19:18:34 |