February 25, 2003
Lazyweb time: Tim, I want to build exactly this in PHP, navigating over a Groove files tool with remote Web Services. Easy, right?
February 24, 2003
What surprised me, which Google was part of, is that superficial search techniques over large bodies of stuff could get you what you wanted. I grew up in the AI tradition, where you have a complete conceptual model, and the information retrieval tradition, where you have complex vectors of key terms and Boolean queries. The idea that you can index billions of pages and look for a word and get what you want is quite a trick.Part of an ongoing series from the Ubiquity magazine.
February 23, 2003
Richard Summers, ex-Agora, likely to have some interesting things to talk about. Hi, Rich!
February 22, 2003
Thank you for the kind words. And, even more, for your seemingly boundless enthusiasm. The future might arrive from different directions, and we need your navigation skills to get to good places.
February 21, 2003
Complementary to the private domain, which certainly is under pressure: the public domain, also at a critical juncture.
Get inspired. How about this for a goal: "universal access to all human knowledge". Public access to the public domain. Is that crazy? Brewster Kahle argues no, indeed: we ought, can, may, and possibly will.
This is over an hour streaming video, but worth every minute. There's nothing on TV anyway, right? If you're really time-strapped, jump in to 46:10, or 54:00 (say). But you'll still miss the awe-inspiring bookmobile and much more. And the scandal of thomas.
More of the series might be worth a look sometime, too!
(I found this on Jon Schull's weblog. But it took a while to realize he'd cut-and-patseyed a paragraph from boingboing. Go follow that link, you'll see what I mean. It's not in any way misattribution, but I don't much like it!).
Ten things to know about XDocs
My take from reading about this app:
February 20, 2003
Fascinating on many levels: Accidental Privacy Spills: James Grimmelmann dissects at length the story of what happened when a reporter sent some interesting but very informal conference notes to a few friends.
Remember how everyone keeps saying that distance is irrelevant on the Internet? Well, this is what happens when distance disappears. You wind up right next to the damndest people. The problem isn't that you can hear your loud neighbors; the problem is that they can hear you.I have two issues with the analysis:  there's a conflation of Internet and Web; and  I don't believe there's a "complete dissolution of the category of the 'private'".
Of course everything is "just bits", copy-and-pasteable to some extent, but we can build environments like your Inbox in which "private" is virtually indistinguishable from "public", and also tools like Groove in which the private sphere reigns by default. Thus also my fascination with Groove-Web interop, in which the technology to blur that boundary proves quite easily tractable, but hints to support and reinforce the appropriate social norms need to be present too.
February 18, 2003
MSFT has shipped a kiddee-Groove, a file-sharing/IM/collaboration tech aimed at teens, called Softie. The project sounds kind of neat, until you realize that it's got an assload of DRM built into it and, in the end, does less than Napster did.It really does sound Groove-like in some ways. Musicmix is a familiar idea, but could use a twist (as Patrick knows - go play).
Robert Faletra in CRN, talking about Microsoft and Real-Time Communication (RTC):
Microsoft ought to leverage Ozzie more or even make Groove the lead dog here. A concerted effort to articulate a product road map while leveraging Microsoft's value-added channel would be a winning combinationHmmmm.
Personally, I'm fascinated. I see the PlaceWare acquisition as a sign that someone understands the need to [a] establish a beach-head with a focused application area (meetingware), and [b] have something with which to define the category of "realtime application server" - that has to include SIP and SIMPLE, but also have broader capabilities.
Placeware has been interesting since the start, because its heritage includes the grandaddy realtime-appserver of them all, one I spent way too much time using:
PlaceWare got its start in 1990 at Xerox Palo Alto Research Center (PARC). In its first iterations, it was a ground-breaking multi-user game known as LambdaMOO.I don't know zip about SIP, but here's one way to look further. Groove is to SharePoint (decentralized application:::centralized application server) as Groove is to PlaceWare (decentralized application:::centralized application server) too...?!
Access control scalability
If you look at the access control problem, it is a matrix. The rows are requestors and the columns are objects of their desire. Linear growth in either or both means geometric growth in the number of table entries in that matrix... beyond a certain size no one can get their head around it anymore... An awful lot of legacy systems' durability is explicitly derived from this combination of "essential" and "cannot be understood."Lovely. Yet, decentralisation of the security can help here too. Decentralization makes things personal and immediate. In the context of security, is it possible to decentralize - make individuals at the edge partially responsible for who they can share information with, by putting some of that directly in their field of view - but keep a sensible amount of stringency, planning and oversight? I think, actually, it is.
February 17, 2003
Matt seems to get the wrong end of more than one stick at once (SharePoint PS is kinda big-and-punchy, and costs, where SharePoint TS is almost a great weblogging tool, and is kinda free) when he writes
a guy named Anil Dash... wrote an article back in October on why Sharepoint is a crappy weblogging tool. He goes through lots of points about why you might not want to use it for weblogs. DUDE! Sharepoint is not a weblogging tool! I am sure if we decided to write one, it would totally rock...I'm tickled by his later entries too - sounds a lot like the sort of thing we used to say about Notes: integrated software, it'll cost ya less to run. Not untrue, mind, but I don't think that'll make many converts from (say) MovableType.
The real deal is something more like "k-logging". Weblog-style publishing as a consciously-instigated business groupware practice. Of which, more later, methinks.
Oh, the luxury
Running a weblog is such a luxurious, inessential privilege - something the Pyra deal will likely change. Meanwhile I get to opine at length, gripe when nobody recognises the hidden brilliance of my latest opus, hallucinate about the size of my readership, and all that. Heh.
So, here's to some of the unsung people who deserve their day in the sun. I'm lucky to work with some incredibly good people - you probably know some of them, like Paresh and Michael and Matt, but not many many others who built this great thing, and taught me a lot along the way. Congratulations. I'm honoured to be a contributor.
February 16, 2003
InfoWorld reviews Groove 2.5
Jon Udell's comprehensive review of Groove 2.5 for the Infoworld Test Center.
Groove founder Ray Ozzie and his teams have always pretended to build application software. But what they have actually delivered are the operating systems of the future — years ahead of schedule.Here's the piece which I particularly want to expand on: interoperability. In particular the SharePoint Team Services glue, which debuts in Groove 2.5 (and I hope will expand going forward). For once, I don't think Jon completely "gets it".
Also new in the Professional Edition of 2.5 is two-way synchronization between a Groove shared space and a Microsoft SharePoint Team Services (STS) Web site. Using a separately licensed Groove toolset called the Groove Mobile Workspace for SharePoint, this bit of integration will appear more seamless to the STS user than it will to the Groove user.Yes, true; we deliberately tried to keep much of the SharePoint user experience in moving STS lists into Groove shared spaces. But even this - lists, discussions - is an education in process.
What do people use to collaborate? #1: Documents. Opaque-blob, resolutely uncollaborative, determinedly non-semantically-queryable documents. Word, Excel, that stuff. #2: E-mail. #3: Discussions. Threaded responses (but this is really pushing the envelope). Documents are the stock-in-trade, and the most important piece of the STS integration because everyone just wants to share (and publish) files.
I work at Groove because I want to change that.
At the opposite extreme to document sharing, we've some Groove platform capabilities and tools which really open the door for "contextual collaboration". The platform's presence and notification mechanisms run deeply through the whole user experience. Then, for example, Pinboard very subtly shows each user what the other people are looking at right now. The experimental MeetCam lets you talk and work together while seeing others' expressions. A proof-of-concept GPS tool lets groups work (on documents, natch) in a space with "place" - showing the real positions of people and things. Several engineering visualization tools let teams work on, say, a support issue, but work with a shared perspective into a 3D model.
The "work on" will usually end up in something for publication: Word, PowerPoint, Web, whatever. Mostly "just files".
The "work in" and "work with" help people get things done, and here, blobs are not enough. Threaded discussions are a good place to start, especially when in SharePoint you can very easily add, say, a "Issue" field which is a lookup into your list of top issues. Then a "Red Flag" field which marks it as a task, or an action item. And so on. The collaborative environment is shaped by its users; it's ad-hoc application design.
Try that with the Groove SharePoint toolset. It just works. And - I'm incredibly excited by this - it's only the beginning.
February 12, 2003
Great word. Here's the canonical definition, if you like.
New from Virtual Methods: Groovelets, a very neat set of small Groove tools. Nice!
In case you were wondering...
Tim Knip has been busy, building a Radio Userland / Groove "interop tool". This little gem uses Groove Web Services (GWS) from Radio, letting you: pull weblog entries into Groove discussions, post Groove discussion entries and files to your weblog, and more. The Groovelounge is an example of what can be done with this integration: a group of people sharing a space, reading news there together, and effortlessly publishing some of their work to the Web.
Fix Error Handling First
Fixing code is easiest when you can run it, and error handling code is only run when there are errors happening for it to handle. Broken error handling code is just a future mysterious accident waiting to happen. Fix error handling code before fixing errors.
February 11, 2003
Happy Fifth Birthday
It's five years since XML was published as a W3C Recommendation. Awesome.
[ /* more blogroll to follow */ ]
The views expressed on this weblog are mine alone and do not necessarily reflect the views of my employer.