This is how you can (and could) find plugins
690 views
Skip to first unread message
Mat
Apr 25, 2017, 7:21:44 AM4/25/17
to TiddlyWiki
In another thread, fellow Josiah asked the folowing. Rather than hijacking that thread I'm replying here:
....
Regarding my own plugins, I should get my TWaddle site active again, to list them. (I've started working a little on this.)
Generally, I hope people know about Erwan's TiddlyWiki Community Search. It allows you to search all tiddlers on TWs that have been "reported" to it and it performs an automated daily update to show which tiddlers (plugins etc) that have been updated. BUT, again, it can only scan TWs that have been 'reported' to it or TWs that are listed in a "root wiki", i.e a TW that links to other TWs. As evidenced, even if it scans daily (and it does), the reporting bit is a bottleneck. (For example, I don't have it in mind when creating my own plugins, which are on separate tiddlyspot domains, and while TWaddle IS a reported root wiki it is currently passive, so...)
Also, ironically, the Community Search project suffers from the very problem it aims to solve: It is not obvious how one should get informed about its existence.
...now what we really need is a system where ones own wiki performs such a scan. And somehow you could get recommendations, via your scans, for other wikis or plugins or whatever.
Yes, TWederation.
Let me add that we have the pieces for this in place and it works. It is, however, still not polished and it is currently too slow to be practical.
For those that don't know, here's a quick run thought on how it could be designed:
How to find plugins / tiddlers / whole TWs
Similar to how the native plugin library works, there can be a default feature in standard TW to "Fetch"... i.e to scans the wikis you "subscribe" to, to get you tiddlers or other information. You decide to subscribe to, and you decide what tiddlers to actually fetch from the TWs you subscribe to.
How do you know what wikis you can subscribe to? How do you find them?
The main route is via wikis you already subscribe to, i.e other peoples wikis that in turn subscribe to other wikis and so you can peek on their lists. That is a "quality stamp". Or someone even has a tiddler with @<yourname> and recommends stuff for you.
But how did anyone find any wikis to begin with? Well, that's easy because it is only cruciual for kicking things off. We could have a listing on tiddlywiki.com or even in the discussion forum.
The key here is information that is created out of individuals personal incentive for quality in their own TWs. You care about which plugins you have installed. This is "stamp of quality", and I'm curious to see which those plugins are. Everyone has an incentive to curate their plugins or really all of their tiddlers. This is in stark contrast to rely on single individuals efforts to keep some external list curated.
And the TWederation plugin itself (again, think of the existing "Plugin Library") could come with a "recommended-subscriptions-list" and this is kept up to date by... you guessed it; fetching. (And who curates that list? It's too detailed to go into that here but, trust me, it's not a problem.)
One limitation is that you can only subscribe/fetch from TWs that are online.
True, but many are. Especially if we're talking about TWs where people present their plugins.
The result is an infrastructure that doesn't rely on a single individual to keep track of everything but instead aggregates small tidbits that several people make about several issues.
It is also an infrastructure that would easify development of the infrastructure itself. You can easily be informed about plugins that enhance TWederation!
...that is the concept of TWederation.
<:-)
....
Mat, could you write a plugin that records plugins?
Regarding my own plugins, I should get my TWaddle site active again, to list them. (I've started working a little on this.)
Generally, I hope people know about Erwan's TiddlyWiki Community Search. It allows you to search all tiddlers on TWs that have been "reported" to it and it performs an automated daily update to show which tiddlers (plugins etc) that have been updated. BUT, again, it can only scan TWs that have been 'reported' to it or TWs that are listed in a "root wiki", i.e a TW that links to other TWs. As evidenced, even if it scans daily (and it does), the reporting bit is a bottleneck. (For example, I don't have it in mind when creating my own plugins, which are on separate tiddlyspot domains, and while TWaddle IS a reported root wiki it is currently passive, so...)
Also, ironically, the Community Search project suffers from the very problem it aims to solve: It is not obvious how one should get informed about its existence.
...now what we really need is a system where ones own wiki performs such a scan. And somehow you could get recommendations, via your scans, for other wikis or plugins or whatever.
Yes, TWederation.
Let me add that we have the pieces for this in place and it works. It is, however, still not polished and it is currently too slow to be practical.
For those that don't know, here's a quick run thought on how it could be designed:
How to find plugins / tiddlers / whole TWs
Similar to how the native plugin library works, there can be a default feature in standard TW to "Fetch"... i.e to scans the wikis you "subscribe" to, to get you tiddlers or other information. You decide to subscribe to, and you decide what tiddlers to actually fetch from the TWs you subscribe to.
How do you know what wikis you can subscribe to? How do you find them?
The main route is via wikis you already subscribe to, i.e other peoples wikis that in turn subscribe to other wikis and so you can peek on their lists. That is a "quality stamp". Or someone even has a tiddler with @<yourname> and recommends stuff for you.
But how did anyone find any wikis to begin with? Well, that's easy because it is only cruciual for kicking things off. We could have a listing on tiddlywiki.com or even in the discussion forum.
The key here is information that is created out of individuals personal incentive for quality in their own TWs. You care about which plugins you have installed. This is "stamp of quality", and I'm curious to see which those plugins are. Everyone has an incentive to curate their plugins or really all of their tiddlers. This is in stark contrast to rely on single individuals efforts to keep some external list curated.
And the TWederation plugin itself (again, think of the existing "Plugin Library") could come with a "recommended-subscriptions-list" and this is kept up to date by... you guessed it; fetching. (And who curates that list? It's too detailed to go into that here but, trust me, it's not a problem.)
One limitation is that you can only subscribe/fetch from TWs that are online.
True, but many are. Especially if we're talking about TWs where people present their plugins.
The result is an infrastructure that doesn't rely on a single individual to keep track of everything but instead aggregates small tidbits that several people make about several issues.
It is also an infrastructure that would easify development of the infrastructure itself. You can easily be informed about plugins that enhance TWederation!
...that is the concept of TWederation.
<:-)
@TiddlyTweeter
Apr 27, 2017, 12:00:37 PM4/27/17
to tiddl...@googlegroups.com
Ciao Mat
I had noticed Erwan's Community Search. But I never used it much because (1) its very incomplete, (2) way behind now, (3) has lots of links that go nowhere (deceased links, old).
What I did NOT know till I read your post was that folk with online stuff need to do a couple of things to get it to track their work. In other words it is tardy simply because so few people are signed onto it.
My 1st question is basic: DO PEOPLE WHO ARE DEVELOPING STUFF EVEN KNOW THIS MECHANISM EXISTS?
My 2nd question is: Why are we not doing more to ACTIVELY PROMOTE such tracking of developments in TW Erwan's thing already makes possible?
My 3rd question is: Why are you not using it :-)
I will reply to your other points later.
Best wishes
Josiah
Mat wrote:
I had noticed Erwan's Community Search. But I never used it much because (1) its very incomplete, (2) way behind now, (3) has lots of links that go nowhere (deceased links, old).
What I did NOT know till I read your post was that folk with online stuff need to do a couple of things to get it to track their work. In other words it is tardy simply because so few people are signed onto it.
My 1st question is basic: DO PEOPLE WHO ARE DEVELOPING STUFF EVEN KNOW THIS MECHANISM EXISTS?
My 2nd question is: Why are we not doing more to ACTIVELY PROMOTE such tracking of developments in TW Erwan's thing already makes possible?
My 3rd question is: Why are you not using it :-)
I will reply to your other points later.
Best wishes
Josiah
Mat wrote:
... I hope people know about Erwan's TiddlyWiki Community Search. It allows you to search all tiddlers on TWs that have been "reported" to it and it performs an automated daily update to show which tiddlers (plugins etc) that have been updated. BUT, again, it can only scan TWs that have been 'reported' to it ...
@TiddlyTweeter
Apr 27, 2017, 12:13:55 PM4/27/17
to TiddlyWiki
Ciao Mat
@TiddlyTweeter wrote:
My 3rd question is: Why are you not using it :-)
Actually, I see you are, though it ain't listing anything of the last 24 hours, and a lot longer too :-)
Best wishes
Josiah
Mat
Apr 27, 2017, 12:55:38 PM4/27/17
to TiddlyWiki
Josiah,
That is why the Community Search is not enough per se. It aims to help people find stuff but if people don't even find the Community Search... no cigar. This is why I keep coming back to TWederation. That is a system that builds on personal incentives and social interaction. It is a LOT easier to add notes if I do it for my own sake than it is to actively share them. For example, just the fact that you have plugins X, Y and Z installed are of potential value, if others knew. You have those plugins because of personal incentives. If you also happen to, say, blog for others to read, I think this is also more likely if you do it in your own TW than e.g here on the boards where you are not in control of your text.
You ask "Do developers know about the Community Search?". Not unless anybody told them. And there is not enough personal incentive to do so, or even to keep the Community Search in mind. (for example, you knew about it but still don't talk about it... not out of evilness but I'm guessing your're of course not thinking of it much.) Besides, few people have as much time as you and I to write on boards like this. Definitely a luxury.
Other than the boards, I'd say people only "read" tiddlywiki.com. But the Community Search is presented on tiddlywiki.com just like any other community site/plugin/whatever. It is not "promoted" or referred to in any other tiddler that I know of. So, my guess is newcomers simply don't get to know about it. Jeremy is the gatekeeper to tiddlywiki .com and he is doing a fantastic job, but he is only one man and IMO simply adding a tiddler for each reported site is not enough. There is no categorization or context to find the info in. And the links are not updated. Besides, his super skills are better used on advanced stuff.
...so again, TWedreation, driven by "the masses" is my best shot for an answer to this information-spreading problem.
I try to bring it up on the discussion forum every now and then (like now). But, there is also some mental resistance to do this because the Community Search is not designed the way I would personally have preferred it. The biggest problem is that is appears outdated and the HUGE thing that it actually updates EVERY day is too subtle.
But, yeah, you knew about it. You're one of the more active fellas here. And still you're not actively promoting it. Why?
As noted, I typically create a new site (a tiddlyspot) for each new creation. And one has to report every site to the Community Search OR report one "main site" into which one writes the other urls. I do have my "blog" TWaddle reported... but TWaddle became passive a year or two ago (because there was no good mechanism to save only public/ready posts. I recently solved this and I'm fiddling aournd with a revamp for TWaddle.) so reporting each new thingy I make is... ufff...
TWederation is also the only thing I can think of where each person himself decides how the data is to be presented, since it is in their own TWs. "I want to prioritize all fetched posts from Josiah", "I want to delete all posts not starred", etc. "I want an alert every time I fetch tiddlers containing the word plugin". Can't do that here or on tiddlywiki.com.
<:-)
I had noticed Erwan's Community Search. But I never used it much because (1) its very incomplete, (2) way behind now, (3) has lots of links that go nowhere (deceased links, old).
What I did NOT know till I read your post was that folk with online stuff need to do a couple of things to get it to track their work. In other words it is tardy simply because so few people are signed onto it.
My 1st question is basic: DO PEOPLE WHO ARE DEVELOPING STUFF EVEN KNOW THIS MECHANISM EXISTS?
That is why the Community Search is not enough per se. It aims to help people find stuff but if people don't even find the Community Search... no cigar. This is why I keep coming back to TWederation. That is a system that builds on personal incentives and social interaction. It is a LOT easier to add notes if I do it for my own sake than it is to actively share them. For example, just the fact that you have plugins X, Y and Z installed are of potential value, if others knew. You have those plugins because of personal incentives. If you also happen to, say, blog for others to read, I think this is also more likely if you do it in your own TW than e.g here on the boards where you are not in control of your text.
You ask "Do developers know about the Community Search?". Not unless anybody told them. And there is not enough personal incentive to do so, or even to keep the Community Search in mind. (for example, you knew about it but still don't talk about it... not out of evilness but I'm guessing your're of course not thinking of it much.) Besides, few people have as much time as you and I to write on boards like this. Definitely a luxury.
Other than the boards, I'd say people only "read" tiddlywiki.com. But the Community Search is presented on tiddlywiki.com just like any other community site/plugin/whatever. It is not "promoted" or referred to in any other tiddler that I know of. So, my guess is newcomers simply don't get to know about it. Jeremy is the gatekeeper to tiddlywiki .com and he is doing a fantastic job, but he is only one man and IMO simply adding a tiddler for each reported site is not enough. There is no categorization or context to find the info in. And the links are not updated. Besides, his super skills are better used on advanced stuff.
...so again, TWedreation, driven by "the masses" is my best shot for an answer to this information-spreading problem.
My 2nd question is: Why are we not doing more to ACTIVELY PROMOTE such tracking of developments in TW Erwan's thing already makes possible?
I try to bring it up on the discussion forum every now and then (like now). But, there is also some mental resistance to do this because the Community Search is not designed the way I would personally have preferred it. The biggest problem is that is appears outdated and the HUGE thing that it actually updates EVERY day is too subtle.
But, yeah, you knew about it. You're one of the more active fellas here. And still you're not actively promoting it. Why?
My 3rd question is: Why are you not using it :-)
As noted, I typically create a new site (a tiddlyspot) for each new creation. And one has to report every site to the Community Search OR report one "main site" into which one writes the other urls. I do have my "blog" TWaddle reported... but TWaddle became passive a year or two ago (because there was no good mechanism to save only public/ready posts. I recently solved this and I'm fiddling aournd with a revamp for TWaddle.) so reporting each new thingy I make is... ufff...
TWederation is also the only thing I can think of where each person himself decides how the data is to be presented, since it is in their own TWs. "I want to prioritize all fetched posts from Josiah", "I want to delete all posts not starred", etc. "I want an alert every time I fetch tiddlers containing the word plugin". Can't do that here or on tiddlywiki.com.
<:-)
Mark S.
Apr 27, 2017, 2:05:59 PM4/27/17
to TiddlyWiki
Instead of another unrecognized 3rd-party system, why not have a page that lists and describes Plugins at TiddlyWiki.com? Similar to the community listings except consolidated in one page. The individual description tiddlers could follow a standard format so people could copy and submit (possibly right to this forum) new plugins.
Mark
Mark
Mat
Apr 27, 2017, 3:34:29 PM4/27/17
to TiddlyWiki
Mark S. wrote:
Instead of another unrecognized 3rd-party system, why not have a page that lists and describes Plugins at TiddlyWiki.com? Similar to the community listings except consolidated in one page. The individual description tiddlers could follow a standard format so people could copy and submit (possibly right to this forum) new plugins.
I'm a bit unsure of the term "unrecognized 3rd-party system" but if it is TWederation you refer to, it could probably be a native feature in TW if good enough. (Jeremy made a prototype for it, different from Jeds, during last years TW European Meeting so I think he is positive to it.) Its "core features" is/could be not all different from the plugin library were you download what you want. Kind of.
But a "page with a list" still has the same problem as the many earlier efforts: Even if people report plugins, who would maintain that list? Update it. Clean out non working ones. Possibly for years. This is one of the things that makes Jeremys efforts so incredible; it's not that the lists on tiddlywiki.com are superb (because they're not! Much because of us not reporting plugins) but that all other attempts by community members to make plugin repositories, recommended lists, etc have simply withered.
Plus, a plugin list is a one-trick-pony. Admittedly, this thread is about "how to find plugins" but the general problem is of course how people should find (and find out about) everything concerning TW.
But, hey, these are just my pessimistic interpretations. Anyone is welcome to set up any solution. I definitely hope it would solve the problems!
<:-)
@TiddlyTweeter
May 3, 2017, 8:38:01 AM5/3/17
to TiddlyWiki
Ciao Mat
There is much in your reply. You articulated ONE thing spectacularly well ...
There is much in your reply. You articulated ONE thing spectacularly well ...
but the general problem is of course how people should find (and find out about) everything concerning TW.
I agree the issue is not plugins per se. Finding them is merely an instance of that broader issue.
IF I can get my brain to coordinate the needed clarity on this I will start another post from the NEEDED starting point.
Best wishes
Josiah
Danielo Rodríguez
May 4, 2017, 9:46:36 AM5/4/17
to TiddlyWiki
Hello,
Sorry but I disagree with all of you.
Please take a look at https://github.com/Jermolene/TiddlyWiki5/issues/2817 to se my motivations.
I think the best place to put tiddlywiki plugins is npm. Just think about it, how do you install tiddlywiki on node? from npm, them, where will probably be your next stop to look for plugins? Probably npm too. Also it will have a positive impact if you just search for tiddlywiki on npm and see the package and a large list of plugins available, that are just one `npm install pluginname` away. So this will also give greater visibility to the project
Lots of projects use this approach: yeoman generators, react plugins, angular directives, babel plugins, gulp plugins, webpack plugins... it is even used for non Javascript code: typescript definitions, cordova plugins
I'm not sure why, but on tiddlywiki community there is a is a trend towards re-inventing the wheel, creating everything from the ground up. I don't see the problem on using battle tested industry wide accepted solutions. And having tiddlywiki plugins on a fantastic highly scalable delivery mechanism like npm is something good.
Obviously this is not for the users that access tiddlywiki from tiddlywiki.com, but for all people that uses the node-js version this will be the desired way.
Mark S.
May 4, 2017, 11:13:36 AM5/4/17
to TiddlyWiki
I mean that there is too many resources scattered in too many places. Having a list right at TiddlyWiki.com would mean that people didn't have to go to a forum to find out that they can go to a site and so forth.
With a simple list, people could submit their plugins either to the forum directly to git-hub. There would be a standard format for the tiddlers so they would automatically get listed correctly. Any person could submit. Since it's part of TW, it wouldn't get orphaned.
Alternatively, there already is a plugin download tool in TW for officially recognized plugins. Why not have a second settings tab that lists all known plugins for download with the caveat that they have not been tested. This is the approach taken by Firefox and most browsers and many applications (like Calibre). You can look for plugins right inside the app. You don't have to find a site or forum, or wade through posts, or ask for help -- it's all right there available straight from the app.
Thanks,
Mark
With a simple list, people could submit their plugins either to the forum directly to git-hub. There would be a standard format for the tiddlers so they would automatically get listed correctly. Any person could submit. Since it's part of TW, it wouldn't get orphaned.
Alternatively, there already is a plugin download tool in TW for officially recognized plugins. Why not have a second settings tab that lists all known plugins for download with the caveat that they have not been tested. This is the approach taken by Firefox and most browsers and many applications (like Calibre). You can look for plugins right inside the app. You don't have to find a site or forum, or wade through posts, or ask for help -- it's all right there available straight from the app.
Thanks,
Mark
Mat
May 4, 2017, 11:16:58 AM5/4/17
to TiddlyWiki
I think the best place to put tiddlywiki plugins is npm.
[...]
Obviously this is not for the users that access tiddlywiki from tiddlywiki.com, but for all people that uses the node-js version this will be the desired way.
That thing is one definite restriction. So it would be not be a solution for vanilla TW. ...Or could it be made into one?
But, more, questions arise in my head:
Who has / what is the incentive to add plugins there?
Who has / what is the incentive to maintain it, e.g remove dead stuff
Is there a front UI to it, ideally in TW, that is made for mortals who simply want a frickin' wiki for their food recipes without having anything to do with "noje.js, npm, github, command windows, ..."
But, more, questions arise in my head:
Who has / what is the incentive to add plugins there?
Who has / what is the incentive to maintain it, e.g remove dead stuff
Is there a front UI to it, ideally in TW, that is made for mortals who simply want a frickin' wiki for their food recipes without having anything to do with "noje.js, npm, github, command windows, ..."
I'm not sure why, but on tiddlywiki community there is a is a trend towards re-inventing the wheel, creating everything from the ground up. I don't see the problem on using battle tested industry wide accepted solutions. And having tiddlywiki plugins on a fantastic highly scalable delivery mechanism like npm is something good.
As hinted, I think it is much a matter of UI and target audience. At lest in this forum, it seems to me that most people are not coders. And IMO they shouldn't have to be, to use TW. I must assume that is the point with WikiText. So things are re-invented but from a TW perspective and probably from ignorance of what already exists.
<:-)
Mark S.
May 4, 2017, 11:17:21 AM5/4/17
to TiddlyWiki
Not everyone uses npm, and I suspect most people at least start out using TW from TiddlyWiki.com. That's why it makes sense to have everything available from TiddlyWiki.com.
Unless ... you're suggesting that TW become a full-time npm program, dropping the single-file, platform-independent format that made it so compelling in the first place.
Thanks,
Mark
Unless ... you're suggesting that TW become a full-time npm program, dropping the single-file, platform-independent format that made it so compelling in the first place.
Thanks,
Mark
Reaktor Blue
May 4, 2017, 12:07:19 PM5/4/17
to TiddlyWiki
Out of curiosity, wasn't the entire "selling point" if you will with TW was that it was one file and portable? Setting this up on node is neat and a great learning experience but the more it seems TW goes in the direction of being hosted via node, the more sense it seems to replace it with an actual mean stack application or something.
Perhaps I'm missing something altogether.
A M Alfaro
May 4, 2017, 12:13:05 PM5/4/17
to TiddlyWiki
I like the idea of integrating a plugin discovery/search in the single file TW because that will always be my preferred platform. In the meantime, though, why not just add something to the group descriptions, especially in the TiddlyWikiDocs group, that says, "Hey! General Tiddlywiki documentation can be found Here, Here, and of course in Tiddlywiki itself, here! To start looking for Tiddlywiki plugins try Here."
I think most non-technical users are far more likely to go to Google and search key phrases like "Tiddlywiki documentation' and "Tiddlywiki plugins", so if the Tiddlywiki groups have those key phrases and signposts in their greeting/descriptions, it'll be very helpful right now.
Just my thoughts,
ama
codacoder...@outlook.com
May 4, 2017, 1:20:32 PM5/4/17
to TiddlyWiki
It should be a sticky post/thread at the top of this group. I'm imagining a single maintained posting, if that can be arranged (otherwise, it'll become too lengthy/tedious to browse/use).
Ste Wilson
May 5, 2017, 5:49:29 PM5/5/17
to TiddlyWiki
There is a fairly comprehensive list on the Reddit group but I'm not sure if anyone stepped up to moderate it after riz said he was stepping down.
Birthe C
May 5, 2017, 7:29:02 PM5/5/17
to TiddlyWiki
Hi Mat,
How would you use twederation for this? I realise it would be a new version of twederation.
No matter what, we need to get twederated. When we get there, we will see lots of new ideas and possibilities.
Birthe
How would you use twederation for this? I realise it would be a new version of twederation.
No matter what, we need to get twederated. When we get there, we will see lots of new ideas and possibilities.
Birthe
@TiddlyTweeter
May 5, 2017, 7:40:55 PM5/5/17
to TiddlyWiki
What Riz did was awesome. Such is pace its already out of date. We need a SYSTEM that keeps up.
Josiah
Josiah
@TiddlyTweeter
May 5, 2017, 7:47:42 PM5/5/17
to TiddlyWiki
We need aggregation.
Nice if Twederation could do it. But its not here. I'm not sure if its anywhere close yet. Lets not put off till tomorrow too much.
I am UNCLEAR what could solve this pressing issue in the near needed future.
Josiah
Nice if Twederation could do it. But its not here. I'm not sure if its anywhere close yet. Lets not put off till tomorrow too much.
I am UNCLEAR what could solve this pressing issue in the near needed future.
Josiah
Mat
May 6, 2017, 9:39:10 AM5/6/17
to TiddlyWiki
Hi Birthe,
How would you use twederation for this? I realise it would be a new version of twederation.
Actually, not much different from the TWederation you've tried. (Hopefully faster fetching tho ;-) But which aspects are you asking about in particular that are not elaborated on, albeit briefly, in my post here above? They key is simply that one is able to batch-wise get filtered sets of tiddlers from other wikis - and once they're in your own wiki, you can do what you want with them.
Further, given this functionality, it would be fairly simple to set up public topic-specialized wikis that fetch and list all, say, plugins (perhaps only meta-data, like links to the plugins rather than the actual plugins), or a documentation specialized with (in this case possibly including the content), or all tiddlers concerning football or what-have-you.
<:-)
Birthe C
May 6, 2017, 1:27:39 PM5/6/17
to TiddlyWiki
Hi Mat,
I wonder how it will cope with many users. I think it would be best for public topic-specialized wikis to fetch links to plugins, as you suggest. I recall the problems we had with TWclassic. Lots of modified versions of the same plugins around. The good part is that most users will get to know TWederation. I think many people are collecting links, writing tips and tricks for themselves to remember and reuse later on. It is done in native language, and the way we ourselves understand it. The quality will vary of course, but it can be shared. Beginners and simple users like myself are the ones needing it the most, and everybody would be able to find stuff and start creating stuff for sharing. We decide ourselves what we want to share, and what we want to fetch, but some kind of evaluation system would be good.
Any thoughts on structure? The TWederation we tried used twCards for identification. Nobody would like to fetch everything from everyone. Would we need lots of different twCards? I realise filters can be used, but it needs to be easy enough for a beginner from the start. It really has to be easy to start finding the topic-specialized wikis, From there it will spread like rings in water.
Sorry to be dense but I do not understand everything in English.
Birthe
I wonder how it will cope with many users. I think it would be best for public topic-specialized wikis to fetch links to plugins, as you suggest. I recall the problems we had with TWclassic. Lots of modified versions of the same plugins around. The good part is that most users will get to know TWederation. I think many people are collecting links, writing tips and tricks for themselves to remember and reuse later on. It is done in native language, and the way we ourselves understand it. The quality will vary of course, but it can be shared. Beginners and simple users like myself are the ones needing it the most, and everybody would be able to find stuff and start creating stuff for sharing. We decide ourselves what we want to share, and what we want to fetch, but some kind of evaluation system would be good.
Any thoughts on structure? The TWederation we tried used twCards for identification. Nobody would like to fetch everything from everyone. Would we need lots of different twCards? I realise filters can be used, but it needs to be easy enough for a beginner from the start. It really has to be easy to start finding the topic-specialized wikis, From there it will spread like rings in water.
Sorry to be dense but I do not understand everything in English.
Birthe
Mat
May 7, 2017, 3:42:43 AM5/7/17
to TiddlyWiki
Hi again Birthe,
Yes, we'll need a way to keep track of versions. If I recall, this is already in place with a kind of UUID both in Jeds and Jeremys variants. But I think that you'll only fetch from a few sources which will work as some kind of quality system. Also, if I recall my conversations with Jed, we talked about trackability and how to make the original source also be part of the meta-info in a tiddler. It's something we'd have to experiment with.
First, much of what I say about TWederation is just my vision of things. Of course, much of it already is possible even with the current TWederation version, but would still have to be built, like a "plugin list". This also means that how things will/can work is up to the individual federated networks and their participants. The transferring fetch-mechanism is one thing but what people build based from it is another, just like TW at large.
We would need some way of identifying sources so, yes, twCards make sense. As for finding the topic specialized wikis, there are a few imaginable variants. The simplest is probably that a few are already build in, just like the plugin library in vanilla TW has the source build in.
Heh! It sure seems like you understand everything.
<:-)
I wonder how it will cope with many users. I think it would be best for public topic-specialized wikis to fetch links to plugins, as you suggest. I recall the problems we had with TWclassic. Lots of modified versions of the same plugins around. The good part is that most users will get to know TWederation. I think many people are collecting links, writing tips and tricks for themselves to remember and reuse later on. It is done in native language, and the way we ourselves understand it. The quality will vary of course, but it can be shared. Beginners and simple users like myself are the ones needing it the most, and everybody would be able to find stuff and start creating stuff for sharing. We decide ourselves what we want to share, and what we want to fetch, but some kind of evaluation system would be good.
Yes, we'll need a way to keep track of versions. If I recall, this is already in place with a kind of UUID both in Jeds and Jeremys variants. But I think that you'll only fetch from a few sources which will work as some kind of quality system. Also, if I recall my conversations with Jed, we talked about trackability and how to make the original source also be part of the meta-info in a tiddler. It's something we'd have to experiment with.
Any thoughts on structure? The TWederation we tried used twCards for identification. Nobody would like to fetch everything from everyone. Would we need lots of different twCards? I realise filters can be used, but it needs to be easy enough for a beginner from the start. It really has to be easy to start finding the topic-specialized wikis, From there it will spread like rings in water.
First, much of what I say about TWederation is just my vision of things. Of course, much of it already is possible even with the current TWederation version, but would still have to be built, like a "plugin list". This also means that how things will/can work is up to the individual federated networks and their participants. The transferring fetch-mechanism is one thing but what people build based from it is another, just like TW at large.
We would need some way of identifying sources so, yes, twCards make sense. As for finding the topic specialized wikis, there are a few imaginable variants. The simplest is probably that a few are already build in, just like the plugin library in vanilla TW has the source build in.
Sorry to be dense but I do not understand everything in English.
Heh! It sure seems like you understand everything.
<:-)
Jed Carty
May 7, 2017, 4:31:19 AM5/7/17
to TiddlyWiki
For twederation the plan moving forward is to make a lightweight distributed server architecture that works like a wiki but has a real database that can act like a wiki. That may be able to help with the scaling problems. The servers would be federated themselves and it shouldn't change the way that twederation works now, it would just be another method for connecting the wikis.
This is still in the theory stages, but there are models of this sort of federation that we can use as models.
@TiddlyTweeter
May 8, 2017, 8:25:10 AM5/8/17
to TiddlyWiki
Ciao Mat, Jed & tutti
Basically here we are talking about aggregation of one form or another to aid knowing what is available
I very much grasp that Twederation (TWED) may well be RIGHT bees knees longer term to do that.
One thing I was thinking about was a universal system of fields or tags that would enable an aggregator to harvest stuff already to some kind of standard. Fully compatible with TWED (even as is). So that other approaches could also work.
There are many issues to sort. I want to make a few points now ...
1 - If TWED were the way to go there will STILL, I think, be need for lists of resources outwith it. TWED will function within TW. BUT there is also a need for resources lists outside TW. Mark S. Has made good points about that. You can't expect beginners to begin unless they know what is available. And its a pain-in-the-ass having to figure stuff out as it is, even for The Constant Reader. Perhaps TWED can be used to generate the needed lists? BUT what should be done before its really working as a matter of course?
2 - Erwin's existing solution to registered stuff is nice. Though falls way behind because few, including people who understand it, register for it. Perché?
3 - I think a LOT of our issues are because Google Groups are so crap. Its fine for immediate interaction. Its terrible for searching and finding past posts (posts that carry the links & works we are talking about). In short, I feel we are positing a savior system when the basic day-to-day system we use is inadequate to the job (job=EASY HISTORY lookup). THAT is making everything more complex.
4 - The whole discussion is bringing up underlying core themes in TiddlyWiki. Some of it is, IMO, an individualism too far. Fragmentation of information is pretty extreme as things currently stand.
Best wishes
Josiah
Basically here we are talking about aggregation of one form or another to aid knowing what is available
I very much grasp that Twederation (TWED) may well be RIGHT bees knees longer term to do that.
One thing I was thinking about was a universal system of fields or tags that would enable an aggregator to harvest stuff already to some kind of standard. Fully compatible with TWED (even as is). So that other approaches could also work.
There are many issues to sort. I want to make a few points now ...
1 - If TWED were the way to go there will STILL, I think, be need for lists of resources outwith it. TWED will function within TW. BUT there is also a need for resources lists outside TW. Mark S. Has made good points about that. You can't expect beginners to begin unless they know what is available. And its a pain-in-the-ass having to figure stuff out as it is, even for The Constant Reader. Perhaps TWED can be used to generate the needed lists? BUT what should be done before its really working as a matter of course?
2 - Erwin's existing solution to registered stuff is nice. Though falls way behind because few, including people who understand it, register for it. Perché?
3 - I think a LOT of our issues are because Google Groups are so crap. Its fine for immediate interaction. Its terrible for searching and finding past posts (posts that carry the links & works we are talking about). In short, I feel we are positing a savior system when the basic day-to-day system we use is inadequate to the job (job=EASY HISTORY lookup). THAT is making everything more complex.
4 - The whole discussion is bringing up underlying core themes in TiddlyWiki. Some of it is, IMO, an individualism too far. Fragmentation of information is pretty extreme as things currently stand.
Best wishes
Josiah
Danielo Rodríguez
May 8, 2017, 12:58:59 PM5/8/17
to TiddlyWiki
I think that the best place that everyone agree is the plugin tab of the control panel. Having some kind of beta tab/channel community driven.
However, until mr. Ruston allows us such thing we are alone here.
From my side, I will start including all the plugin libraries that gave me their authorization into the official distribution of NoteSelf.
Regards
Jeremy Ruston
May 8, 2017, 2:12:28 PM5/8/17
to tiddl...@googlegroups.com
Hi Danielo
I think that the best place that everyone agree is the plugin tab of the control panel. Having some kind of beta tab/channel community driven.However, until mr. Ruston allows us such thing we are alone here.
Really? What is it that I am not allowing? Can you point to a statement from me that backs up your assertion?
Best wishes
Jeremy
From my side, I will start including all the plugin libraries that gave me their authorization into the official distribution of NoteSelf.Regards
--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at https://groups.google.com/group/tiddlywiki.
To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/ab40d213-53ee-4b5f-9616-bf83f9ebc135%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
HC Haase
May 9, 2017, 3:23:14 AM5/9/17
to TiddlyWiki
I agree with Danielo Rodríguez and Mark S. that some tab or extra plugin library in the regular plugin modal, is the best way to make a plugin findable (for now).
A couple of people have made there own plugin libraries (e.g. Tobias).
Could this be a solution:
1. Have a TW on github where people can add tiddler files (.tid) like the way you can add to documentation for tiddlywiki.com at the moment.
2. Have that wiki publish a plugin library (if that is possible at the moment) that is included in TW (e.g. community plugin library).
3. Have plugin developers upload there plugins there by adding the .tid file (and using the tinka plugion builder or something similar to get the plugins in the same structure).
This would give
1. one place in the wiki to find community plugins.
2. maintenance and publication of plugins would be on the individual plugin author, so less of a bottleneck/ work for admins.
if plugins was made using the tinka plugin things like version number and TW-verision comparability, would make this library more easy to manurer for the user/plugin-consumer.
would this be a solution technically? and one we would use/want?
A couple of people have made there own plugin libraries (e.g. Tobias).
Could this be a solution:
1. Have a TW on github where people can add tiddler files (.tid) like the way you can add to documentation for tiddlywiki.com at the moment.
2. Have that wiki publish a plugin library (if that is possible at the moment) that is included in TW (e.g. community plugin library).
3. Have plugin developers upload there plugins there by adding the .tid file (and using the tinka plugion builder or something similar to get the plugins in the same structure).
This would give
1. one place in the wiki to find community plugins.
2. maintenance and publication of plugins would be on the individual plugin author, so less of a bottleneck/ work for admins.
if plugins was made using the tinka plugin things like version number and TW-verision comparability, would make this library more easy to manurer for the user/plugin-consumer.
would this be a solution technically? and one we would use/want?
@TiddlyTweeter
May 11, 2017, 6:45:42 AM5/11/17
to TiddlyWiki
We are talking about an aggregated repository for info & links on plugins & stuff, right?
Maybe my brain is too small but isn't the issue, at this level, simply, to encourage folk to use Erwan's automated mechanism that already exists!
result: http://tiddlywiki.com/prerelease/#Latest
signing-up: Set up for Community News
Its seems it would, at least in the interim, go a long way to addressing the issues NOW in a way that is good enough until TWederation, or some other solution, is universally in place.
Best wishes
Josiah
Maybe my brain is too small but isn't the issue, at this level, simply, to encourage folk to use Erwan's automated mechanism that already exists!
result: http://tiddlywiki.com/prerelease/#Latest
signing-up: Set up for Community News
Its seems it would, at least in the interim, go a long way to addressing the issues NOW in a way that is good enough until TWederation, or some other solution, is universally in place.
Best wishes
Josiah
HC Haase intelligently wrote:
I agree with Danielo Rodríguez and Mark S. that some tab or extra plugin library in the regular plugin modal, is the best way to make a plugin findable (for now).
<snip>
Danielo Rodríguez
May 16, 2017, 11:51:05 AM5/16/17
to TiddlyWiki
Hello Jeremy
Really? What is it that I am not allowing? Can you point to a statement from me that backs up your assertion
Sorry if I looked impolite. I can't find any particular source to make a quote, but I asked a couple of times how to make a plugin fit into the official plugin library and the answer was something like: "That's Jeremy's decision"
Which, to be honest, is a true statement.
Please note that my sentence:
until mr. Ruston allows us such thing we are alone here.
means that you are the owner of tiddlywiki.com and tiddlywiki's repo, therefore a decision like including a beta channel on the plugin library should be supported and approved by you. Again, sorry if this is not true, but I have the impression that you don't want to include such feature. Am I wrong ? If so I would love to contribute to such thing.
Regards
Christopher Arcadia
Jun 7, 2017, 11:35:47 PM6/7/17
to tiddl...@googlegroups.com
After several years of using TiddlyWiki (without many plugins), I just found Tobias' huge Tiddlywiki Classic plugin stash on GitHub and cannot help but think there are many other 3rd party plugins that I would love to check out!
I think the most accessible approach to helping TiddlyWiki users explore its capabilities through plugins would be to have the plugin search bar (in the control panel of each wiki) also search a repository for approved 3rd party plugins.
I think the most accessible approach to helping TiddlyWiki users explore its capabilities through plugins would be to have the plugin search bar (in the control panel of each wiki) also search a repository for approved 3rd party plugins.
Reply all
Reply to author
Forward
0 new messages