The Table Ruler: A List Apart - this is
exactly what I was thinking about less than a week ago. Nice to see that someone's already made it, so I won't have to write it from scratch :-)
ODP now indexing RSS feeds - wow. Actually, I
don't think this is a good idea, at least not the way that they're planning on doing it. Today, virtually all (100%?) RSS feeds contain content that is basically the equivalent of an HTML page. It wouldn't make sense to inlude my blog
and my blog's RSS feed as separate listings in the directory. What
would make sense, however, is if
along with my blog, they listed its feed. Otherwise they're just going to have lots of listings that are basically duplicates.
Headline Index - HTML - Canada's
Globe and Mail now offers RSS feeds. I've subscribed to one, but I'll probably end up unsubscribing. Nothing wrong with the Globe and Mail, which I like, it's just that traditional media has nothing to offer me (directly) anymore. Via
Will Pate. Btw, they could improve by offering more specific RSS feeds (search results!).
Anil Dash: unsolicited advice - “...It doesn't matter if you thought of it first, or if your idea is better. If you didn't do anything with it, if you didn't push it, then don't come complaining...” via
Anil.
My earth lab today was a field trip. We visited a bunch of places including a landfill, recycling depot, and sewage treatment plant. We ran out of time so didn't see the water treatment plant. I was writing and listening most of the time, but below are two photos from the landfill, taken from inside the bus. Apparently birders come to the landfill to count the seagulls; makes sense. The sewage treatment plant surprisingly didn't smell as bad as you'd expect it to. The talk by the guy giving us a tour of the landfill and adjacent projects mentioned quite a bit about the economics of it. Quite interesting. Methane from the landfill is used to generate electricity; we had to wear earplugs as we went through the generating station.
change is inevitable. the point is to direct it
Inspired by some of my recent posts, here's a (very incomplete list) of what you need to do to be a
With-It Company, mostly general, but some specifics, some of which don't apply to everyone
- open and communicative
- tell people what you're up to, what you're working on, what you're planning on. [specific: publish a blog (with RSS feed) and/or newsletter, even a wiki perhaps]. communicate directly with customers and other interested parties, and use their feedback. admit your weaknesses; don't pretend your competitors don't exist, acknowledge them.
- no lock-in
- make it really easy for customers of your competitors to be customers of yours [specific: easy data import]. make it easy for people not a customer of anyone to be a customer of yours. make it easy for customers of yours to be customers of your competitors [specific: easy data export]. even better, allow your customers to also be customers of your competitors at the same time: provide integration [specific example: any search engine can be added to the Google deskbar]
- always in motion
- continuous small improvement. let people know you're alive and working [specific: blog!]
- other stuff
- that's all on my mind at the moment, maybe I'll add more later...
I have just finished reading
The Visual Display of Quantitative Information (Amazon link) (
Wikipedia link) by
Edward Tufte, and I'm reading another of his books now (both from the university’s library). I was
first lead on to this author by
Luigi earlier this month, and I'm certainly glad about that. There's a lot of things in the book that I was already inclined to, and also some new things. Things I want to remember:
- integrate text, tables, and graphics; display tables and graphics inline; use more text labels on graphics
- turn non-data-ink (ink in the metaphorical sense, these days) into data-ink, such as by using the axis to show ranges, distributions, and/or real values
- avoid using more dimentions to display data than is in the data; i.e. area to show linear data, and if actually doing so, scale properly
- never use pie graphs – this one is interesting, and I will have to think about it more
- graphics can cater to multiple views, i.e. overall, horizontal, vertical; information immediately obvious, with more revealed as you look at it in other ways
- a lot of data can be fit into quite a small area, viewers are intelligent; small amounts of data are usually better in a table than a graphic
Fagan Finder > Weblogs, Journals, and RSS - this is an archive of my blog search page, as it existed in
August 2002 until the next time I updated it. It included "Search for Blogs" (9 tools), "Search in Blogs" (2 tools), "Search in RSS Headlines" (1 tool), "Search for RSS Feeds" (2 tools).
14 total, all in English.
Today: "Search in Blogs" (8 tools), "Search in RSS Feeds" (15 tools), "Search for Blogs" (19 tools), "Search for RSS Feeds" (8 tools), and "Metadata collectors" (2 tools).
Total 42 tools, including six non-English ones. In addition to the huge increase in blogosphere search tools, there is the massive increase in tools for searching within RSS (and Atom) feeds.
GUIdebook: Graphical User Interface gallery - insanely comprehensive gallery of screenshots of various GUIs for various things from various operating systems. Via
Blogdex. My general stance: systems should be easy to learn if you've already used a similar system, but better :-). That is, generally, don't redo everything, unless what you're doing is very revolutionary. Or at least have both traditional and new ways to accomplish the same tasks.
Websites have multiple audiences. There's the power users, the occasional visitors, the visit-once-for-info-and-leave people, the visiting-for-the-first-time-may-become-power/occasional-user people, and the people who came looking for something that's not there. There is no reason why website's shouldn't adapt to different users, and in some cases, they should. I don't feel like giving examples now.
Viewpoint ViewBar Introduction - haven't really played around with it yet, but it looks like this toolbar is actually
not just a clone of every other toolbar. all hail innovation
XML.com: The Beauty of REST [Mar. 17, 2004] - a book tool has been on my agenda for some time. Perhaps I'll be able to get around to it in a couple of months. Jon Udell has already done some of what I plan on doing. Maybe some cooperation shall occur. Also, it's an example (certainly no offense to Jon), about how the simple things often are the most popular. Via
Pilgrim.
Landslides in the News - the Canadian government's website includes a "news feed" powered by RocketInfo, a Canadian company. I wonder if RSS is involved?
And yes, I am working on my assignment at 5:10am, and the assignment is due at 11:30am.
Update: guy from Rocketinfo e-mails me. Now this is a company that is on: using Feedster queries to keep up with little guys like me talking about their company. Anyhow, when I asked about RSS, I meant the format travelling from Rocketinfo to Natural Resources Canada. The answer is that it travels in XML, which I presume isn't RSS, but that doesn't matter. And this is a company-client relationship, by which I mean that the Canadian government agency is paying for the feed. Rocketinfo sees part of a business model here, and they're right. Also (naturally), they suggested I take a look at their web-based RSS aggregator.
I skipped my calculus class today to attend a talk by Jack Layton. For those non-Canadian readers, he’s the leader of the New Democratic Party of Canada, left-leaning, with basically no chance of winning the next election (they now hold 14 of the 301 House of Commons seats).
The turnout was bigger than I've seen for other events, although really it should have been bigger. As the guy sitting beside me said, how often do you get to hear a talk by a federal party leader? I only found out about the talk two days ago, and none of the several people I bothered about it came with me. Bah.
Anyhow, the talk was quite interesting. He began by talking about climate change. He actually used the term WalMart-ization. He talked about regulating industry to encourage cleanliness and energy effiecieny, and discourage pollution, about funding transportation. He wants hybrid cars to be manufactured in Ontario (that's the province I live in, the most populous in Canada, with lots of vehicle manufacturing). He commented that we could be at the forefront, not way back like Bush. He also criticized the Liberal government and Paul Martin (the Prime Minister), including that Martin wants to decrease Canada's debt, yet just proposed that the maximum student debt be increased. In that example, as with others, he pointed out that it is the Banks that benefit from that (and his tax cuts) which in turn fund the Liberal Party. Apparently Sweden doesn't have tuition; Jack considers education a right, not something that should be part of the market economy. He also pointed out the conflict of interest of Paul Martin's company (now his kids' company) making money by shipping coal. Another big issue was the US's STAR WARS program, which he points out would probably not work anyway. Mention on decriminalizing (or legalizing, I don't remember, check their webite) marijuana, along with a joke about that. He got a few good laughs during his talk. Mentioned that young people aren't voting (very true), and how important that is. His position on gay marriage (yes) got an ovation, as did quite a few other points. Of course, I guess that isn't unusual, since the crowd was probably mostly supporters. Talked briefly about how racial profiling and pesticides are bad (not related, of course).
Overall, it was much more interesting that my calculus class would have been. I wonder why at every event there are people who take 10,000 years to ask a question; that really doesn't show respect for others who have questions, when the speaker has a limited time. There was a CTV news cameraman there.
I am subscribed to the RSS feed from Feedster of search results for things happening in Kitchener and Waterloo, where I'm currently living. About two weeks ago, one of the results was a press release, Sterling-Hoffman announces Angel Mehta to speak at University of Waterloo. I live on campus, and it looked interesting, so I RSVPd; it meant arranging with one of my professors to attend a different class, but that wasn't a big issue.
I thought that I might be the only first-year student there, or the only non-computer student, but I wasn't, on either count. I guess there were around 20 people all together. I was there early, and was talking to people before it began. One of the organizers was asking how we had found out about the event, and my answer was definitely the most interesting. I didn’t try to explain RSS or anything though. Another organizer didn't even seem to know about the press release. And I did meet someone who knew about RSS and Feedster. Would you believe that I go to a school famous for computers, and I think this is the first person I met who knew what I was talking about?
Anyhow, the talk was definitely interesting, especially the real-world anecdotes. I cleaned out my pocket- and wallet-suppy of business cards, I should have brought more with me. One woman (a recruiter) was highly impressed with whatever it was I told her, which I guess would be from the business card and website, and being "only" 19. Looks like I'll be attending their next event on the 31st, which will be more participatory I think.
Query the Canadian National Earthquake Database (NEDB) - on a related topic, I was copying data from this tool into Excel, and auto-separating the data by delimiters. The "date" column I naturally told Excel was filled with dates. The interesting thing is that all of the dates before 1900 didn't work at all.
Update: not surprisingly,
others have noticed the problem and found solutions.
City of Ottawa - I've previously written about how
university course websites suck, but the same applies to university websites as a whole, and pretty much any (yes, there are plenty of exceptions) government-run or government-sponsored websites, including those of
libraries. I’m doing a project for my Earth Science class on Ottawa; naturally I looked at the official Ottawa (it's the capital of Canada, btw) website. I ran the English home page through the
Semantic data extractor and it I was shocked to see the sum total of all extracted data:
the title.
News - gaim - installed Gaim (instant messaging software) yesterday. Don't think I'll be going back. Also, I've got a Jabber account now.
CeBIT to premiere USB Swiss Army Knife - I was almost laughing a this... until it occured to me that I carry both a USB memory stick and a swiss army knife on me at all times ;-). This item will sell.
Facts on archiving, search and retrieval - interesting. and also illustrating my poor (self-chosen) list of topics. This will make less sense if reading via my RSS feed which is (now) not showing topics.
HubLog: How To Find (More Of) What You Want - remember when I used to blog all the comments I madeon to my sideblog? I really need to get around to working on a solution to this. I should dedicate three weeks when I've got time, and spend it lobbying people like
Six Apart.
And while I'm on that topic, I want to add a new extension to FOAF. Something like "services used". So I could write something like (bear in mind that I know virtually nothing of RDF, namespaces, etc.)
<foaf:Person>
<su:service su:name="del.icio.us">
<su:account>mfagan</su:account>
</su:service>
</foaf:person>
Update- obviously I should have checked the FOAF spec before proposing an extension. What I proposed seems to bascially be there already. I need to take a closer look.
Ask Jeeves to buy Interactive Search for $343 mln - this list (showing ownership of selected web property aquisitions) is for my own reference. I think it’s all correct.
Yahoo!: Overture (incl. AlltheWeb, AltaVista), Inktomi, various non-search sites
Ask Jeeves - Teoma,
Direct Hit, Interactive Search (incl iWon, Excite, myWay)
LookSmart: Zeal, WiseNut, Grub
AOL: Netscape (incl. Open Directory), SingingFish
Terra Lycos: HotWired (incl. HotBot), various non-search sites
Google: DejaNews, Sprinks, Applied Semantics, Pyra (incl. Blogger), a number of tiny companies
MSN: nothing
InfoSpace: Go2Net (incl. Dogpile, MetaCrawler), WebCrawler
Primedia (About.com): nothing
Fish! - pictures of a wide variety of interesting looking sea creatures. Via
Blogdex.