Right now we have the listing page, libreho.st
but it’s difficult to find out what services are hosted by what hosters, All that page offers is a list of librehosters.
Which is good and better than what there was when there was nothing, but one improvement would be to have some kind of table, so each hoster puts, on their entry, what they offer, so there is a yes or a not or a ‘soon’ to things like 'offers … email? lists? vservers? nextcloud?
For that, we would need to extend the directory format to add services
And that should be prioritised
It should go in the json so we can display it. Just we need to add it to the format.
I would also propose, if possible, to order them a bit less randomly. I think I would prefer to put first the hosters that do not invoice and only accept donations, and below, and if possible in a different colour. the hosters that can and do issue invoices that can be shown to tax men, government instances, because they pay their taxes and their workers.
(right now there is only the addition order in reverse: the newer the entry, the closer to the top.)
I would prefer a random order (which I think it was). I don’t think discriminating on the economic model is sound because 1) people need to pay for the resources and make a living, 2) “only accepting donations” can hide an economic model where your funding source is not necessarily a better option than paying customers.
I agree that the available services list could be mentioned somewhere via the librehosters API.
Why do you think donation based is the best for a hosting project? Or even better, what is your ordering algorithm, and how do you justify it? I’m genuinely curious
There are many ways for ordering, and random seems like the easiest that we will all agree
This is a really good idea. The ontology we are currently using is based on https://schema.org/ so there is already everything that we need and even more.
So currently each hoster is an organization.
An organization can have an offerCatalog that is basically a list of offer and I think we would agree that we all offer services with serviceType webApplication.
Maybe we’ll need to add InternetApplication
I’d agree that the ontology is really business… And does not really embed the diversity of our group. We can also PullRequest Schema.org to add our specifies, but it is some work.
I think it would be better to use an already existing ontology. If you know another one that you think is better, great.
If you think it is a no go with schema.org and there are no better, then, I’d be fine if we create our own. But I’d prefer to be sure that there are no other better ones first. (and that you don’t like schema.org, I personally can live with it )
I’d agree that the ontology is really business… And does not really
embed the diversity of our group. We can also PullRequest Schema.org http://Schema.org to add our specifies, but it is some work.
I think it would be better to use an already existing ontology. If you
know another one that you think is better, great.
No i do not. In fact I am not familiar with anything in the quoted paragraph
If you think it is a no go with schema.orghttp://schema.org and
there are no better, then, I’d be fine if we create our own. But I’d
prefer to be sure that there are no other better ones first. (and that
you don’t like schema.orghttp://schema.org, I personally can live
with it )
ofcourse feel free to develop and share the results or viceversa
I would prefer a random order (which I think it was).
It looks more an inverse chronological order; the older ones are at the
bottom.
I don’t think discriminating on the economic model is sound because
I don’t think I am proposing to discriminate; just ordering them using
some criteria so that users can find what they look for more easily and
the economic model is an important variable when looking. itis likeif
you are looking for a person by name; if a listing is separated in men
and women you dismiss half the list right at the start so it can be easier.
so I think the conversation does not necessarily come into this
direction but I’ll answer anyway, if that is ok …
people need to pay for the resources and make a living,
in the case of volunteer collectives we only need to pay for the
resources and making a living is done elsewhere
“only accepting donations” can hide an economic model where your
funding source is not necessarily a better option than paying customers.
I dont think by accepting donations we hide anything … the rest is
philosophy and again I don’t think it addresses my proposal, I am not
saying one is better than the other I am only proposing separating the
two so that projects’ economic models dont seem hidden
anap:
For that, we would need to extend the directory format to add services
And that should be prioritised
Sorry @anap, I did not mean you (or your collective) in particular. At P.S.: we do not ask for anything, because we have funding and do not want to hold more responsibility than we can handle. I was more thinking about the current discourse (e.g., at FOSDEM) where a volunteer community was simply co-opted by the GMAFIA almost without a fight. I don’t think it’s the case with Librehosters, but I think it’s good to clarify the point.
In order to extend the directory format (proposal), the change should happen in the librehosters/librehost-api repository. @anap, please open an issue referencing this discussion, with your proposal for new fields. I see that there are already 4 open issues, maybe we could collectively have a look at them :]
[…]I was more thinking […] where a volunteer community was simply
co-opted by the GMAFIA almost without a fight.
It is not the first time I see this issue mentioned as a big problem. It
may be that where I hang out there are not volunteer communities strong
enough to atract GMAFIA
I don’t think it’s the case with Librehosters, but I think it’s good
to clarify the point.
