InfiniTiddly (or, Rich Shumaker is evil) ;-)

[David Gifford]

Hi all

Rich Shumaker is evil! ;-)

Thanks to this thread https://groups.google.com/forum/#!topic/tiddlywiki/AatW5wFls-w, I discovered Rich Shumaker's "TW inside a TW" http://t5a.tiddlyspot.com/#_canonical_uri%20Example. That got me thinking again about TW as an infinitely expandable content management system. So thanks to Rich I spent a day tinkering instead of getting my own work done. So he is only evil in that he distracted me from my work for the day. But actually I am grateful to him for doing that! Thanks Rich!

I created a central hub file called InfiniTiddly, then I made a number of adjustments to the Psalms TiddlyWiki I wrote about earlier today and created a portal to it in InfiniTiddly using Rich's ObjectFrame.

http://giffmex.org/notes/infinitiddly.html

The Psalms tiddler found there is actually a portal to access the Psalms TiddlyWiki. I could fill Infinitiddly with objectframes to many other TiddlyWiki files in this way, and have an fairly infinite library of tiddlers accessible in one TW, and a user might not necessarily realize that that is what is happening. Filesize becomes irrelevant.

I am happy with the results so far, but there are some wrinkles I would like to see ironed out before I start using this widescale:

On the bright side:

1. Viewing: I hid the sidebar in the Psalms TW so that what comes into view in the Infinitiddly is the default tiddler, not the sidebar. The Psalms file has a white background now so that it does not contrast with the tiddler contents.
2. Printing: When printing from Infinitiddly, only the tiddler contents shows, not the icons or either of the frames.
3. At the top of the viewtemplate in the Psalms file, I have a home button, a download/save button, a link to a special search tiddler, the chevrons to reveal the sidebar, and more/edit/close. They are off to the left so they won't interfere with the outer InfiniTiddly more/edit/close, and they are separated by color to make it easier for the eye to scan and memorize them.

On the medium side: There are probably more icons there than most people will want to stomach: 3 icons for the frame tiddler in Infinitiddly, and seven icons in the tiddlers at the Psalms file. In particular it would take some getting used to, to avoid closing the outer frame tiddler when what you want to do is close the inner Psalms tiddler. But I am content with this for my own use, and I think it looks pretty clean. You be the judge.

Problematic and beyond my limits:

1. While on my laptop (Firefox in Windows) long tiddlers have scrollbars, on my iPad they do not, and there is no way to get to longer content.
2. When printing longer tiddlers from Infinitiddly, the Psalms content below the bottom of the Infinitiddly portal tiddler is cut off.
3. There is no way to permalink a tiddler from the Psalms file while viewing from the Infinitiddly file. The Infinitiddly file is what stays in the browser address bar.
4. The process of viewing the content from the Psalms tiddlywiki is pretty slow on the iPad.

Would love your suggestions: Are there ways to fix these four problems? Are there other problems with these files, blind spots I have not considered? Are there easier and more appropriate ways to do the same thing (I had to tinker with several core shadow tiddlers, and the CSS is probably a mess).

Blessings,

Dave

[David Gifford]

Bump.

Hi guys, I always forget not to post on weekends. I never get any feedback from anyone when I do. Or maybe people were uncomfortable and didn't understand that my comments about Rich were a good-natured joke, given that his object frame inspired me to use most of my Saturday playing with it.

Any feedback on my use of Rich's object frame in http://giffmex.org/notes/infinitiddly.html or help you can give me on the hurdles I still face (see original post) would be great.

Dave

[Andrew]

If you revisit the page where Rich links to my page you'll find that the object viewer pretty much does exactly the same thing that native external image viewer does with _canonical_uri field. The only reason I created my object viewer was so that I could display pdf files. Since my company converted all procedures to html, I don't think I'll need to use it. Let me be clear though, it is an iframe and it is sandboxed. It is not an exteral wiki includer. Someone still needs to create an Includeplugin.

[David Gifford]

Hi Andrew

1. Oh, so you are the one who created this thing that took up my Saturday, not Rich! :-) Great work!

2. Yes I realize it is an iframe type of animal, not an includeplugin type of animal. Nevertheless, with certain restrictions, the way I have it set up it can be used to access numerous external TWs. I would just like to see a few kinks worked out (or worked around), is all.

3. I hadn't thought about its application for pdfs. Great to know.

Thanks for this wonderful tool. I am loving it.

Dave

On Mon, Feb 23, 2015 at 10:32 PM, Andrew <andrew.j....@gmail.com> wrote:

If you revisit the page where Rich links to my page you'll find that the object viewer pretty much does exactly the same thing that native external image viewer does with _canonical_uri field. The only reason I created my object viewer was so that I could display pdf files. Since my company converted all procedures to html, I don't think I'll need to use it. Let me be clear though, it is an iframe and it is sandboxed. It is not an exteral wiki includer. Someone still needs to create an Includeplugin.

--
You received this message because you are subscribed to a topic in the Google Groups "TiddlyWiki" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/tiddlywiki/wZxqMEtEx4o/unsubscribe.
To unsubscribe from this group and all its topics, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.
For more options, visit https://groups.google.com/d/optout.

[RichShumaker]

I have been off the boards for a bit as you can tell because I am just now responding to this ;)

Thanks Dave for the kind words as I do live backwards, #heheh, and yes good-natured joke back at you.

I have always enjoyed the stuff you do so keep up the great work.

Now my TW of TW's is actually more of a directory of TiddlyWiki's I have created.

My goal is to mash up 3 things on a server(I am using a RasPi right now)

1. TW of TiddlyWiki's - Not embedded or _canonical_uri as Andrew did in the example you list but more like a glorified bookmark list.

2. Spawn the TW's from each other using BidiX store.php clone trick.

3. Use Erwan's Aggregator(Link to Github) for the files on the Server to bring it all together and searchable.

If this mash up worked what would that allow you to do?

You could quickly create new TW's so you could have TW Templates for different things(Notes, Math, Video, Photos, and more) then 'clone' them easily and quickly.

They would automatically show up in your TW of TW's(this part is very manual right now).  You must copy and paste the URL right now no automagic.

Each new TW you create would then be combined automagically into a searchable system every once in a while, TW Aggregator does it at a set time so every 10 minutes as an example.

The TW Aggregator can scrape a local directory and do this.

I envision this being used by a teacher in a school, a hackerspace, or someone like me that has a bunch of TW's.

Currently I am missing a way to automatically add the cloned TW back to the TW of TW's directory.

The rest is already set up and I just need to implement and test it.

Rich Shumaker

[Alex Hough]

Dave,

do we take it that you are "back"

Alex

--
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.

[Jed Carty]

Rich,

That sounds awesome. It sounds like taking the philosophy of tiddlers and applying to wikis themselves where you can have a wiki focused on a very specific subject that is connected to other wikis forming a broader knowledge base. This could greatly simplify sharing large groups of tiddlers because you could share the entire wiki that contains only those tiddlers.

Almost completely off topic, but I got a raspberry pi 2 and might have blown the fuse on it within 3 days of getting it. If it isn't broken I am hoping to set it up as a syncthing server as a dropbox replacement. Your idea seems like something that would work well on a server like that.

[David Gifford]

@Alex, not really, unfortunately! But when I get an idea that will help me with my goals I will share it and/or ask for help/feedback.

@Jed: A knowledge base using a central TW linking to other more specific TWs is something I have been doing a long time, at recursos.giffmex.org in Spanish. But viewing those specific TWs through object frames makes the movement from the central TW to the subTWs more immediate: one doesn't even leave the central TW.

@Rich: Sounds like a great idea. I already do the first of your three steps with my own TW files in English. I had seen your Spawn a tiddlywiki tiddler on Saturday and was intrigued by it, but haven't tried it yet. The third step is nice in that they are all searchable. I will need to think about that some more and experiment at some point if I am up to it. Nice "idea sex" - a phrase I learned today, nice combining of three ideas together to form a new hybrid idea.

But I still want help with my questions/problems in the original post...

Blessings

Dave

--
You received this message because you are subscribed to a topic in the Google Groups "TiddlyWiki" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/tiddlywiki/wZxqMEtEx4o/unsubscribe.
To unsubscribe from this group and all its topics, send an email to tiddlywiki+...@googlegroups.com.

[RichShumaker]

For

3. There is no way to permalink a tiddler from the Psalms file while viewing from the Infinitiddly file. The Infinitiddly file is what stays in the browser address bar.

I don't think there is anything you can do as you are not actually viewing the real TW file you are viewing a "snapshot" of it when you use "_canonical_uri".

I call it a snapshot but it is not an image of the file the real file is there it is just you have no control over the data.

So you can't transculde data or create lists for the data unless you went to the other location and did it there an then look at it through the lens of the _canonical_uri.

It is like viewing a book through glass as you can't actually touch that data directly from inside the TW you put it in.

That was my understanding at least and I will need to try this to confirm but that is how I remember it when I was working with it.

This is why I liked the idea of the TW Aggregator as you actually get all the data merged into a single TW file and that data can be manipulated separately.

Rich Shumaker

[David Gifford]

Hi Rich

Yeah I knew there would be no way to access the data from my central TW. But since you can click links and open tag pills, I thought there might be a way to click a button in a tiddler in the viewed file, and have it execute an action of "grab permalink, and open it in new tab."

Does the TW Aggregator actually expand the filesize of the single TW file when it accesses the data from the other files? Does that file get larger and take longer to download each time?

Dave

--

[Jed Carty]

From what I understand the TW Aggregator gets bigger, but it only takes data like the title, creation date and creator from each tiddler so it shouldn't get too much larger unless you have a huge number of tiddlers in it.

[RichShumaker]

When you use _canonical_uri you are including the entire X or Y.

So if it is an image you get the whole image.

If it is a website you get the website the way it is designed, some dynamically load, TW HTML loads in the entirety.

So if you take 5 TW's and put them all inside a TW using  _canonical_uri then you load 6 TW's.

The TW Aggregator strips out the Tiddlers and puts them in a single TW file.

So yes it is larger but it is not double the size if you use 2 TW's or triple if you use 3.

This is my understanding of how it all works.

Please correct me if I am wrong or if there is something I am missing.

Oh and I hope this helps.

Rich Shumaker

[David Gifford]

Thanks Rich.

But, just making sure, the canonical_uri and the objectframe only add the TW file *while the tiddler is open*, and it disappears when the tiddler is closed, right? So if I had a dozen object frames to one site each, but only opened one at a time, I would not be using 12x the space, right?

About the aggregator: Does the aggregator grab tags? And can one tell the aggregator which fields to include?

Dave

[RichShumaker]

I do believe that canonical_uri loads only when you go to that tiddler.

So if you open one at a time then it loads that data one at a time.

What I don't know is if you then reload after you close the Tiddler or if that is buffered or cached?

Again I know how it works just have not tested to see what goes on behind the scenes.

So does anyone know the details of 'canonical_uri' in action?(how it loads, what it loads, space it takes, time it takes, ect. ect. ect.)

In regards to TW Aggregator I don't think it it 'configurable' per se but it does grab the Tags.

Here is the Github Repo Issues for that project and it may have more details.

I love the TW Aggregator for what it has done so far BUT I have not used it yet as I am still setting things up.

Rich Shumaker

[Tobias Beer]

Hi Dave,

I think, in general, other than for showcasing purposes — but even for that, too — embedding one TiddlyWiki in another ain't a great idea... printing issues being just one aspect. Better use a "hub" or "shared menu" so as to quickly switch contents. If you want an independent menu accross wikis, you may want to consider a plain html file with the links that you embed using an iframe in every wiki.

Better still, use a html file containing a frameset, holding a simplistic menu-html file in one "menu-frame" and then the actual wiki in a content-frame. In other words, clicking on a link in the menu-frame opens the corresponding wiki in the content-frame.

Best wishes, Tobias.