Improving links.tiddlywiki.com

#1

Given that the url collection project (can’t think of the name) that inspired this approach went under, maybe it’s not too much of a surprise if it doesn’t take off here either.

The problem is that for most people the process of adding new items is too arduous. Also, the fact that you can’t just have Links collect it from your already existing TiddlyHost file means you have to have a dedicated TW file just for favorite links. The theory there is that it would put too much load on the server if the links were part of general-use files. My feeling is that the loading process is so fast, and the updates so few, that it wouldn’t actually be a difficulty. Like most people will only update their personal, public TW’s once a week or so. So even though the individual load might be a bit much (by bit much, I mean more than 1 second), it happens so seldom that it doesn’t matter. So, maybe deal with the load problem further down the road if becomes a problem. Right now the problem is the opposite – lack of uptake.

2 Likes
Query: Does https://links.tiddlywiki.com/ have a section for Plugin Libraries?
#2

@Mark_S I think discussion around improving uptake for Links would be better served in its own thread where it will hopefully get proper attention.

2 Likes
#3

@saqimtiaz I can see how we submit links and its quite smart how this is organised. As you suggest

Rather that the community is bookmarking what they find useful or interesting.

However I would like to respectfully suggest https://links.tiddlywiki.com/ leaves a lot to be desired on the UI design. Verbose text, unreadable links, too much whitespace and no keywords or highlights.

As a desktop users, especially when searching for tools and plugins I would prefer maximising use of the screen space.

However I was excited to see the links can be downloaded as a JSON of tiddlers. So I could craft my own interface. This makes me ask perhaps if we could design a simple aggregator (as links is) and then attache an RSS feed type mechanism multiple sites could then subscribe to the links and have alternate presentations of the same aggregated contributions.

See how this extracts the tiddler title from a link, one of many techniques to extract the maximum information and display it.

<$text text={{{ [[https://abesamma.github.io/#Interesting%20links%20about%20TiddlyWiki]split[#]rest[]decodeuri[]] }}}/>

Questions;

  • Can I contribute to an improved interface on the links site, and how?
  • Is a subscription to the aggregator possible?
  • could we allow the source links to contain more information
#4

It was a familiar story: deli.cio.us was acquired by a bigco, which quickly killed it – see Delicious (website) - Wikipedia. But later, Pinboard, another indie bookmarking site with roughly the same functionality acquired the remnants:

https://pinboard.in/

Pinboard is quite popular, at least in some communities. It’s sometimes cited as a good example of a solid, long term subscription model.

I tried to word the warnings about that carefully. As you infer, the server load isn’t a problem right now, and so it is a shame if that concern provides friction to adoption.

Right now, the batch job runs in about 2 minutes:

There’s plenty of headroom there; the tw.com build process runs for 5 minutes or so without problems.

TiddlyWikiLinks is hosted on GitHub:

You can contribute directly on GitHub via issues or PRs, or by raising topics here on talk.tiddlywiki.org

If you mean an RSS/Atom subscription, then no, that’s not there at the moment. But you can subscribe to the JSON data.

At the moment we display the text field of the link tiddlers. What else are you proposing?

#5

Small comment. Yes! And not just because it’s an initial hurdle.

There is also an implicit assumption that you should maintain it that is a possible turn-off.

I’m not so sure flagging an item ace is the same as maintain your view of aceness forever.

I put that crudely. But I’m not entirely clear about what is needed in practice to flag the best resources?

Just a half-formed comment
TT

#6

Right! An issue is that to “bookmark” is usually conceived of as “a click or two”. This is somewhat (a lot) more than that!

What I’m trying to point to is the idea this is simple bookmarking ace stuff kinda fails. It’s an investment of time and effort to do the biz needed.

And, as far as I can see, links.tiidlywiki.com has no facility to UP-vote items.

Just a comment
TT

#7

Hi @TiddlyTweeter

The current system supports links in TiddlyWiki format. Surely the one thing that people in the TiddlyWiki community have in common is that we know how to save things to TiddlyWiki.

I’d be keen to support other sources for reading bookmarks; anything with an open API is feasible.

2 Likes
#8

I’d say, as an end-user, the single biggest issue for me is “taggitus” … submitters use the folksonomies they invent (rightly). The problem with that is we end up with a swamp of tags, many that could be merged to simplify things for end-users.

I have no idea how you could solve that.

Just saying
TT

1 Like
#9

This has come up before: the plan is to have a table of aliases to map the tags
Styles,Stylesheets and Styling to a single standardised tag.

We need to figure out the rules we’ll use for the standard tags (eg pluralisation, sentence case etc). I’ve been working on the tooling in the background (we’ll need a little editor app in TiddlyWiki to make it easier to manage the aliases).

1 Like
#10

There are a few small things that might help:

  • provide via the Links site a simple button to add to your own TW to create new Link tiddlers, ala Saq's links — Useful TW resources
  • an option on a site like TiddlyHost to create your own Links wiki, with a button like the above already set up.

I personally quite like that my TW bookmarks are separate from the smorgasbord of my overall bookmarks.

2 Likes
#11

Ciao @jeremyruston, this is askance but I think relevant in a way. TW is extraordinary in being so w–i–d–e (meaning I’m not so sure it’s comparable to anything else as it is so self-contained-comprehensive).

I think that makes it quite (well, very) difficult to give a fully accurate TOC of it’s doings once you get to all the things people are doing with it that ramify in further multiple directions.

But it is, I think, an issue on uptake that the beginner can get lost in a fog of “too much”.

This is half-formed thought, but drifting towards what I want to say.
TT

#12

The fact that I couldn’t try before buying deterred me from investigating further.

My feeling is that this:

Should be removed until such a time as it becomes important. Likewise this

is actually the opposite of what is wanted. It may not have been a good thing in the world of deli.cio.us for developers to insert their own bookmarks, but in the world of TW it’s a good thing since otherwise the work may not be noted at all. Also, there’s no field, flag, or standardised mechanism to indicate that an author is submitting their own work.

When I reviewed the tags, it appeared to me that condensing the tag names, although a good thing, would only reduce the numbers of tags by maybe 10%. There will still be a lot of tags. I guess that just means that people are using TW for a wide breadth of topics. Which is good. But it does mean a bit of cognitive overload is inevitable.

Also there appears to be a routing bug. If you click on “Plugins (74)”, for instance, it will take you to “plugins (1)”, and only show one tiddler.

I think overall it works pretty well. The main thing is it would help if it had a better search interface, possibly like the advanced search on TW itself.

#13

For me, the ideal work flow would be that I capture bookmarks, and some of them I flag for public links use. Having to post it in two places puts friction in the process. If that’s true for me, then it’s probably even more true for more casual users.

#14

Understood, and do please feel free to move to that model. If the server load becomes an issue we can deal with it, but it seems unlikely to be a problem in the short term.

#15

I tend to have a lot of open browser tags of links from this here place then intermittently splurge on putting them into my bookmark wiki.

I tend to copy the description from the site of there is one, then copy/ paste the url into the correct field. Also need to remember to make the description good enough so that it makes sense without the title as links dosnt present that.

Making up general taggery as I go (you’re welcome).

I think it’s an important thing to do as we create a lot of cool stuff here and @davegifford is a busy man!

I think the ease of use is definetly a barrier. Delecious worked because it had a browser plugin. Press the button, it was added to your bookmarks. I can’t remember if there was a dialogue to enter tags or not. I’ve slept since then…
Pearltrees is my fave current bookmarking solution for similar reasons ( Have a look if anyone needs some engineering and education related links).
(I really should do a tiddlywiki one…)

#16

**peatrees.com** ’s server IP address could not be found.

Spelling?

#17

Corrected. Should have read pearltrees.com.
Cheers for the catch!

#18

Ooh. Thanks for this. But it’s annoying what they do to your username. I typed “DesertDwarf” and they stored it as “desertdwarf” (which is OK), but then they displayed it as “Desertdwarf”. Why be inconsistent? It’s not like where “Desertdwarf” (or “Stephenteacher”) is displayed is part of a sentence. It’s just the username all by itself. As a dwarf, it’s the little things that matter. :slight_smile:

1 Like