The Seafile Community Manual has more than enough content now to be published very soon. There is still a lot of content that has to be edited/added/removed.
Before we can do this I’d like some people with good english (not perfect! ) to proofread and test what the manual asks to do.
Please write me a PM if you want to contribute. It helps a lot if you know how to use GIT or at least know your way arround usage of the Github GUI.
Thank you guys!!
Edit:
This is not the “server manual” from Seafile. The CE manual is mantained and published by community members only. You can read all about it if you contribute or once it’s released.
Update:
Please read this post for new information.
Of course there I am using Github. I don’t want and probably won’t mantain this manual my whole life. So other contributers are welcome to join. But no public link on the forum or people will find it because Google will find and spread it.
I’m late to the game on this, but I’ve encountered several things that are either erroneous, outdated, or incomplete in the manual. I’ve issued one pull request, but I’m new to Git. So, my question is, is a pull request enough to get the manual changed, or is there something else we must do?
Just one, but I only did a pull request. I didn’t edit any pages. I just issued a pull request with the details for the proposed change. I need to familiarize myself with Git. Other than using it to read code, I’ve never done anything else with it excepting that one pull request.
I am working on the Github Repo to make it work for public community availability. I am looking for people that help maintain, check and approve pull requests. Soon I’ll add a linter check that will help to verify that the written content is valid markdown and follows recommendations. It also makes sure that links are valid and other important stuff.
For now please make a fork and open pull requests if you want to change something.
I’d love to have the manual filled with good information and to complete it before we start working to fix syntax errors (there will be quiet a few).
So who is with me and ready to work as a team maintaining a good and well structured CE manual?
I also plan using waffle.io to manage changes to the manual. So we can even make the process of managing this fun.
I still think it would make more sense to help improve the official server documentation. I don’t think a separate one is maintainable. It would also always lack documentation of the newest features and would require a significant amount of time to maintain.
I do not agree. Several members of the community tried this in the past. The Seafile Ltd. team won’t accept any aproaches from members who restructure the whole manual in a way that it’s actually pratically readable. They rather ignore such requests and “don’t have time to check pull requests”. Even normal pull request to update or extend the existing manual are ignored. Pull request just waiting forever to be approved or even read. No thank you. Don’t need that waste of time that I (and I guess others too) invest in this software. Yes it open source but that still doesn’t mean that the documentation needs to be crappy. Seriously. I won’t repeat the same again. Seafile is not taken seriously on the market and the documentation is only one reason why it’s not.
The CE manual covers the CE edition exclusively and how to get it working. It is not supposed to replace the official manual or extend it. And it’s not impossible to maintain.
You (@DerDanilo) bring up an interesting point here: “Seafile is not taken seriously on the market and the documentation is only one reason why it’s not.” Why do you think Seafile is not being taken seriously? Are you speaking about the company, the software, the licensing? Just curious what your thoughts are.
Company (at least in Europe) and therefor the software itself since a lot of companies cannot use it because of the way how the development is handled and how Seafile Ltd. manages updates. Also it’s just sad how new features are released within the pro version without proper and complete excessive testing because “we don
t have the manpower to do that”. This is just an excuse not to do a very important task and lift the quality of this product. Yes it got a little bit better but there is still so much more space for improvement.
Just have a look in the web, there is not much around Seafile as Enterprise Product. There is nearby no marketing and when someone found it they loose interest quiet fast once they find the no go points.
I am not interessted to work with Seafile Ltd. but am fine to invest some of my free time into the open source part of this software and the documentation. Which doesn’t mean I have to do it to the original manual.
Is waffle.io FLOSS or based on something that is FLOSS? I personally don’t want to use any proprietary tools anymore (great proprietary tools are great vendor lockins in the end).
Edit:
Sorry I just saw:
I am working on the Github Repo to make it work for public community availability. I am looking for people that help maintain, check and approve pull requests.
The day M$ bought GitHub was the final reason for me to disable my account there. As explained I don’t want to use any proprietary tools anymore (GitLab would be okay for me, better a Git hosting software which supports federation but I don’t know if such alternatives are already out there).
Idc which one we use but GitHub is still the one most people know and use. And please think about the fact that we are using it to host a community manual and not any program code. Also we could move it to any other Git repo of it should become legally necessary.
If you want to contribute you may create a special email address for that and use it only for GitHub.
Thx, “Gitea” was the one with federation support I didn’t remember at time of writing.
Who is “no one”? If this is the fact why GitLab or Gitea were developed? If people wouldn’t care about self-hosting possibility projects like Seafile would be totally obsolete. The opposite is the case.
What the “most people know and use” can be an indicator to determine if something is “good” but should never be the only one and actually not the most important one.
If software “most pople know and use” would be the only only/most important one I would use:
something like Dropbox instead of Seafile
a “walled garden” system like WhatsApp instead of Email, XMPP and Matrix
Apple/Windows instead of GNU/Linux
Google Android/iOS instead LineageOS
PlayStore instead of F-Droid
(- easy to use, well known non-free/proprietary software instead of sometimes hard to use, less known software that was designed to respect my freedom)
Does this make really a difference? My point is not about what is hosted but where and how it is hosted.
That’s true but again I think this is not the point in what should really matter: Following your argument I could say: Save your data using something like Dropbox instead of Seafile. The reason you use Seafile is (hopefully) not only related to privacy concerns but as well that you are interested in decentral network structures that respect your freedom and the idea of the internet as a decentral global network.
I personally don’t really care too much about privacy (if I really would I shouldn’t use almost any modern computer system because all of them use proprietary parts like baseband chips or management engines no one can audit except the manufacturers).
What I care about is to avoid supporting any centralized structures which want the users lock into different kind of “walled gardens”. I admit that the side effects of walled gardens are less heavy in case of GitHub but the only reason for this is that GitHub uses git which is a free(dom respecting) system. As long as GitHub itself is no FLOSS using it means (for me) that I support a digital service which has no interest on decentral structures and (implicit) no interest on real user freedom.
And yes, thinking this way causes the same problems like deciding not to use Facebook: I exclude myself from the digital majority. Especially for FLOSS projects this is even more hard for me because I can not contribute any bug reports to my favorite projects like matrix.org (unlike Seafile there is no official forum which I could use instead).
I can understand why FLOSS (so freedom focused) projects like Seafile decided to use GitHub as code hosting platform in the past (works well, has a really nice look&feel) but (since FLOSS based platforms like GitLab or Gitea are out there) I can not understand why they still use a non-free platform to develop their free software projects.
Sorry I cannot understand your post at all. Maintaining an own server to host the community manual would probably be more work than so far has gone into the manual itself. Furthermore it would require all users to register, which most won’t do.
There is a clear advantage of using github.
no administration hassle
many developers and users already have an account
it works and provides everything one needs
with travis ci and github pages publication can be automated fast and easy
there are no costs
On the opposite for using gitlab there would be the following disadvantages
it requires a server (monthly costs)
it requires administration
from time to time something breaks and needs to be fixed
most users won’t create an account just for the community manual and won’t contribute
automation would require extra work and extra resources
And I’m already someone who always tries to work as independent as possible, but there are enough cases where it isn’t worth it. This clearly is one.
The reference platform is proprietary and contains many, many, many features gitlab ce does not have.
The point is millions of devs/users already have an account and can use it for thousands of projects. For an own gitlab server they would have to create it for one project.
See above this is not given with gitlab as well. When freely using it one would use features of the enterprise edition very fast and isn’t able to move to a ce server. Furthermore we’re are talking about git. So no matter which software, migrating the repository itself is extremely easy.
Furthermore I don’t think that Richard Stallman thinks Gitlab is floss. There is one company deciding everything (=no freedom) (and also developing most of the stuff) and the software is only partly free.
) to proofread and test what the manual asks to do.
