Monday, December 25, 2006

When I wrote SdiDesk I explicitly decided that I wanted many small, hand-layed-out diagrams which are hyperlinked together, rather than automatically generating a large diagram. (Of course, it helped that that was the lazy thing to do, in terms of coding :-)

Manually drawing diagrams doesn't help with *discovery* of new connections or of other structure but, for me, it's the right way to capture some of the structure in *my* thinking. I tend to create diagrams with around 10 (15 max) nodes where I might be deliberately representing a hierarchy or a supply-chain or some other small chunk of the world. And I can arrange the nodes in a way that makes sense for that kind of thing. If I really need more nodes, I probably want to break it up into an overview page and sub-pages with details. Of course, SdiDesk nets are more primitive than even I want, visually (so yeah, I am thinking about how to improve on that in my next opus) But I prefer them to Buzan-style mind-maps where I find most software forces me to build one "butterfly" shaped hierarchy. Or to something like TouchGraph which is very clever but actually not much use when you want to get acquainted with the shape of your information.

3-D web of idea nodes : TiddlyWiki ... - Smart Disorganized Individuals -

Saturday, September 23, 2006

SocialText 2.0 is out.

Check out the screencast. Some nice work on the UI.

Here's what looks good to me :

* the automatic backlinks

* deciding you can be "expert" (an explicit recognition that SmartAscii mode corresponds to expertise),

* watched pages,

* the sortable table views of RecentChanges, watched pages and other queries. (I like this very much.)

What I'm still ambivalent about :

* "tagging". Why are tags special? Why are they not simply a convention on top of full-text search (as in CategoryCategory)? Maybe it's simple optimization. Faster to search an "index" of explicit tags. Maybe the flickr / folksonomic culture is too deeply ingrained to fight?

* they haven't quite got there yet with blog integration. As I'm always banging on about, the tough problem with wikilogs or blikis is that you have two incompatible addressing schemes : pagenames and dates. Deciding which addressing scheme you need for any particular item of information is difficult. As SocialText 2.0 allows everyone in a group to edit blog-posts as wiki pages, blog-items can blossom into full pages. What's not clear is how you can link to a particular blog-item, or how such an item could migrate out of the context of a specific conversation and become a more independent citizen of the wiki.

As always, the problem isn't technical but one of shared concepts. If you have two types of objects with distinct addressing schemes : temporal and permanent, how do you make sure that you and the rest of the community understand which information is of which type? Are people putting it in the right place? Are they looking for it in the right place?

I don't have any answers. But I'll only really get excited about a wiki that *also* provides blogging when someone finally does find a nice way to help the users navigate this conceptual gap. Otherwise, you might as easily use two different products. (Not trying to beat up on SocialText here. I'm just saying that this in one of the grand unsolved problems in the area, and they don't seem to have solved it yet.)

Why I'm still on the edge of my seat :

* WikiCalc? What's happening here? When will grids become first class citizens of the SocialText product?

Note here that conceptually, grids are much less trouble to integrate than blog-posts. Yes, they're technically very different from ordinary pages. But addressably they're the same kind of animal. Individual cells are perhaps more problematic, although the spreadsheet convention is very strong.

Given that Chris Dent's at SocialText, and EEK's involvement, do they (still / yet) have purple numbers in the SocialText wiki? Perhaps they do but that's too geeky for this screencast.

In fact we may see SocialText having to come to grips with questions about integrating four different addressing schemes : page-names, purple-numbers, cell-addresses and blog-items. I suspect proliferation of incompatible addressing schemes will be one of the major issues faced by wiki as it starts to take over the enterprise.

Thursday, September 07, 2006

Very interesting.

Now HyperScope has been released, don't miss Brad Neuberg's great Screencast of Douglas Engelbart's Augment System

How much was SdiDesk influenced by Engelbart? Directly, not so much. But indirectly, obviously a lot. My direct influences were wiki, the Smalltalk environment, Emacs, HyperCard and the web-browser. Obviously, all, heavily indebted to Engelbart's work.

Wednesday, August 23, 2006

Dave Winer says he made 2.3 million with his blog, without advertising.

What I think he means (as I explain to Bill Seitz, here) is that he sold the server last year.

For Winer, the blog is a full media and communications strategy. It’s how he does his personal branding, hypes his projects, and most importantly makes his connections with important users and developers.

Without Scripting News Dave wouldn’t get invited to the right kinds of parties and conferences, wouldn’t have people working with OPML, or reading River of News, or his podcast directories etc. Without Scripting News how would he (or anyone else) understand that there was even an opportunity called “” or what form it should take?

Update : Somehow this reminds me of another Steve Pavlina article : the $10,000 hour.

Not in the letter, of course, - Pavlina's tactics for commercializing his blog are antithetical to Dave's - but in the spirit.

Both eschew mere optimization of the normal microchunk in favour of the hard-to-measure, riskier, higher potential stroke of genius. For Pavlina that means merely increasing your hourly rate by gradual, evolutionary steps is a bad deal if by doing so you miss out on the $10,000 flash of inspiration. Winer says the same for merely monitizing eyeballs rather than making the connections, and having the conversations, that lead to the next

Sunday, July 23, 2006

WTF!!! I've been Googling for references to SdiDesk, like, forever, but I've never seen August Agricola before today!

Wow! The guy has an SdiDesk mod that looks beautiful.

And sounds like he's done a tonne of cool stuff with it. Damn! I wanna see more!

Good point too, about SourceForge. The stupid thing is I even have an account ... just never got round to uploading it ... doh!

Here's more from August.

Monday, July 03, 2006

SdiServer screengrab
Originally uploaded by interstar.
Don't get too excited. This is just eye-candy at the moment. :-)

But it is what you think it is ... a page from an SdiDesk PageStore served by a web-server (I'm using, and viewed in Firefox with some cute graphic tabs.

Thursday, June 29, 2006

Hey, guess what ...

SdiDesk rocks!!!


I've been working, on and off, on the Python translation and making some progress, but the whole VB code-base had gone stale for me. And, although I've been using SdiDesk for the odd notes, I've been dabbling with other things like the OPML Editor and writing more on online wikis.

But this week I've started a new job in a fairly corporate environment. I've been expected to get up to speed on a lot of things fairly quickly. This is a local division of an international company, selling one component in a larger enterprise system. In the first three days I've had to get an idea of where our place is among various partners, suppliers, customers, systems integrators etc; I've had to get a picture of a range of different products and modules and libraries produced or consumed by different players; I went out yesterday and watched as someone upgraded the installation of our product on-site with one of our partners and will probably be expected to do it by myself the next time. SdiDesk has been absolutely, amazingly, invaluably useful as a place to dump all the information which is being thrown at me.

I've used most of the features : drawn a quick network diagram to sketch out our position in relation to our partners, clients, their clients etc. etc. And as each box in the diagram links to its own page, I can add other information about the parties : web-sites, contact info etc. I've drawn similar network diagrams to show the relations between the various tiers and modules in the software. My first drafts are often wrong but the diagrams are easily modified as new data comes in.

I've written out the steps to installing the software, using nested bullet lists. And, naturally, cross-linked to the modules when relevant. I can throw in login names and passwords. I've used footnotes, and tables, and hyperlinks to both online documentation and the actual web-based interfaces to the system. I'm confident that SdiDesk will become the dashboard that I will work through.

I've also seen its many flaws. SdiDesk has many rough edges, bugs, abandoned and unfinished roads, and is distinctly ugly. But, frankly, I haven't been this excited by, this in love with, it since the release of 0.2. I'm definitely back into it in a major way. (Of course, I don't have any time :-)

So if you don't know what I'm talking about because you haven't yet checked out SdiDesk; but you work in an environment where you need to cope with a lot of information including information that is list, table and graph-shaped; and you're a smart, if a little disorganized, individual; then you know what to do.

Yeah, there are still dozens of things that suck or are aesthetically challenged. Yep, I'm a slow and inattentive developer. But, to misquote Tati Quebra-Barraco, SdiDesk "is ugly but it's fashionable".

(And works off your pendrive.)

Tuesday, April 18, 2006

I notice Eastgate do stationary.

Is this the future for GTD / personal organizer / hyper-text companies? Give away software, sell paper based products?

Saturday, April 01, 2006

Thursday, March 30, 2006

Can you run SdiDeskOnAPenDrive?

Of course.

Hmmm. I wonder if there's a market for a generation of software aimed at pen-drives. As a complement for server-based web 2.0 stuff.

Friday, March 17, 2006

Another wiki-oriented blog : Wiki That!
Brilliant graph of user happines and feature creep. Well worth bearing in mind.
Recent discussion between Albert and me. Albert highlights what SdiDesk does, and what he wants it to do. I give my latest, still evolving plans, for future SdiDesk development. Yes, Python is the future.

(NB : it's a long page that needs refactoring. Scroll down.)
Here's a cool looking blog dedicated to wiki : : Understanding Wikis
Hey! It wasn't me. But someone made an Sdidesk page on Wikipedia. !!! :-)

Monday, March 06, 2006

LOL! Dave Winer on DMOZ vs. Wikipedia

DMOZ has a bad rep for having editors with conflicts of interest. And it's exclusive, unlike Wikipedia which at least has battles (never thought I'd say that) of people with conflicts, DMOZ doesn't even have dissent among conflicted people, only one point of view exists, because there's only one editor for each category.

(My emphasis.)

Scripting News: 3/6/2006

Interesting to note how Dave is spinning against the DMOZ hierarchical catalogue (I never heard of any repution for conflict of interest that he claims, though it's plausible) just as he's about to launch a revolution in online hierarchical directories. :-)

Thursday, February 16, 2006

Stowe Boyd : The Individual Is The New Group -- Part 1

which kind of reminds me of all those ideas about PersonalKnowledgeManagement that were floating around last year.

Monday, January 30, 2006

Quick comment I made about personal timeliness vs. personal individuality Blogger

100% right.

At some point we realize that some of our information consumption has to be a "minority game".

The winners will be those who know things that other people *don't*. That's what makes you valuable to other people. Being an expert in the top ten most known topics makes you worth what?

By definition, there's no algorithm which can give this to you. You need to cultivate your own tastes and ideosyncrasies. Your outlier blogs.

Don't worry about being the first to follow the fashions. Something which is number one on memeorandum or digg today (and is actually important) will probably turn up on one of your regular blogs within a week. So decide how you'll trade-off your own personal timeliness for personal individuality.
This isn't a problem. It's the solution.

Politicians who try to spin their wikipedia entry are getting named and shamed.

Thursday, January 26, 2006

Latest news copied from the ShowStoppers page

I'm still thinking, more and more seriously, about rewriting at least the wiki-to-html engine in Python.

I'm also seriously thinking that the next release will actually be a complementary "web-server" edition, which allows you to edit pages via a browser (over the web) rather than through the VB program.

Right now I have a simple webserver application which let's you access the SdiDesk pagestore via the browser over HTTP.

Of course, this raises questions about security which will have to be answered before release. Will there have to be a login? Can you mark some pages as personal and private, while others are public, etc?

If I go 'browser only' I'll initially lose the network diagrams. (Would anyone miss them?) Although it may be possible to put them back for the more recent Firefox using SVG / Canvas tags.

None of this is ready yet. But it is looking promising. And the main reason is that the new code is in Python which will allow me to work on a proper Python based rendering engine. (Where I should fix all those bugs.)

If the Python rendering engine works out, I'll then try to retrofit it behind the current VB front-end.

Thoughts anyone?

Tuesday, January 03, 2006