« August 2006 | Main | October 2006 »
Tomorrow is Museum Day. What is Museum Day? Started by the Smithsonian Institution, it is a day when a number of museums across the country offer free admission to visitors who present the Museum Day Card available on the page I just linked to. For those in Western PA, the Carnegie is participating, so this is a great opportunity to check it out if you haven't been before.
New Knitty! New Knitty! Curl up with a cup of coffee and plan your projects for the next couple of months! (That's what I'm doing!)
Ivy is a gorgeous wrap sweater with amazing cable details at the waist and cuffs. On the sock front, Red Herring have a nice herringbone pattern and is the pattern from this collection I'm most likely to actually make. Little slip of a thing is a great-looking felted bag that - very cool - only uses one strand of yarn throughout but supposedly still has fairly robust structure. Intolerable cruelty is also drawing my attention, even though it probably shouldn't....
David Brin laments the lack of simple built-in programming environments on personal comptuters [via Slashdot]. I too remember learning to program on my Apple IIe - if you turned on the computer without a programmed disk in the drive, you fell into BASIC, and I copied many listigs out of magaziines or books and played around with their functionality. Brin is entirely right - this type of built-in, no-fuss programming environment got a lot of us started.
Now, there are still command-line options. My programming students download Java off the Sun website and compile and run from the DOS prompt, and they could use Notepad to write their code, though I think a more supportive editor is desirable. But, Java isn't accessible in the same was BASIC was. And installing and running it this way requires some wrangling with your PATH environment variable - particularly if you have Quicktime installed.
And Brin points out that even if you can fairly quickly get a Java environment (or C++, or.....) going on your computer, these alternatives do not match the advantages of BASIC. I'm not going to head down the path of arguing comparative programming languages, though I think there are other programming languages that can be interesting tools for early exposure (okay, I'll just mention LISP and its functional ilk.....) but will agree that our modern robust languages don't lend themselves to back-of-the-book, type-it-in experimentation.
The whole article is a really good read, but Brin's bottom line point is that without this ease of experimentation, today's children will grow up with the computer being a perevasive tool but no more understanding of how it really works than most of my generation has of the workings of our car (especially compared to the knowledge of our grandparents). Says Brin:
The parallel technology of the '70s generation was IT. Not every boomer soldered an Altair from a kit, or mastered the arcana of DBASE. But enough of them did so that we got the Internet and Web. We got Moore's Law and other marvels. We got a chance to ride another great technological wave.
I thtink Brin is a bit too dismissive of the value of "information consumption devices" when engineering and used properly. But given that there is no technological reason why such devices can't also allow easy access to minimal programming environments, I think he is right to question whether we are advancing ourselves beyond a point that invites the energy and enthusiasm of novices and hobbyists.
The latest Craigslist kerfluffle described here at Slashdot reminds me of the case from about a year and a half ago of someone posting their chat sessions with a plagiarist soliciting them for an essay. In both cases, you have people assuming that their one-on-one conversations will be kept private and sharing information with a complete stranger that they would not want made public.
Clearly, this is not a nice thing to do. I think it is also unethical - unlike the plagiarism case where the recipient of the advances was able to check that such behavior was prohibited at the student's school, there is no reason to believe that what these people are doing is wrong, even if they are married. Certainly, it is anti-social. The same "experimental results" about rates and types of responses to an on-line solicitation could have been reported while ensuring the respondants identities.
But, it is unbelievably stupid to not assume this will happen every time you share personal information over the internet. One response has been that this type of revelation of personal information is illegal, citing Washington state law. I hope the problem with this objection is obvious - not everyone on the internet is governed by the same laws - they may be in different states or countries. In short, there is no way to stop this without fundamentally changing the structure of the internet.
It's simple - if you wouldn't say it to your mom, your boss, your best friend, your worst enemy, and your next door neighbor, don't say it on the internet to a complete stranger!
The latest post at India, Ink about using templates in document formatting software made me feel a little guilty, but reminded me why I actually miss LaTeX a little bit. I really don't understand styles and templates in MSWord, which I used to format my most recent couple of papers, and while it is nice to throw together a handout without having to wrestle with LaTeX, I missed its clarity of structure when dealing with sub-sub-sections and captioning figures and including citations (BibTeX, I miss you so....)
The point at that we ought to learn to use our tools is valid. However, I do think a tool like MSWord discourages us from really learning to use it as a document design tool as compared to a document layout tool. All of the things that are made easy or provided with icons have to do with forcing a particular view of a specific piece of text. You can select a paragraph and change its indentation, font or alignment with a single button-click or keyboard shortcut. You can create an entire document in "normal" style, and it looks fine to the eye. But, as India notes, this doesn't scale, and it isn't reusable. It is horrid design.
And yet, while LaTeX puts the design issues front and center, MSWord makes the process more mysterious. Perhaps it can do this, and I have never found the feature, but being able to show the user a marked up version of their text, like the switch between the WYSIWYG and HTML panes in Frontpage, would help, but I've only been able to check my designated style by highlighting the text of interest and seeing what the style menu says it is in. Not good for a quick consistency check across a whole document!
I am picking on MSWord, only because I've been using it for forever, but the criticism holds for the other WYSIWYG text editors I've used as well. They may allow good document design, but they don't make it easy. I suspect that India is working with way more high-powered tools than these, but I think her point stands that we don't worry about the design of our text documents the way we ought to, particularly when there is so much potential for them to change presentation format, and the consumer grade tools ought to address this.
My most recently completed knitting project is this beaded mohair shawl, my first attempt at knitting with beads. I followed the "Patagonian Night Sky Shawl" pattern from Knit and Crochet with Beads by Lily Chen, though I included more repeats than the book called for since I wanted a slightly larger shawl. the yarn was a remaindered machine knitting cone that I got cheap a few years ago and had been looking for a purpose for; the beads are iridescent black glass beads - see the detail photo below.
It turns out that knitting with beads is fairly easy, though you have to be a bit careful about keeping the beads on the correct side of the work. The mohair helped, because it had good friction. But, stringing the beads was a huge pain and I had my yarn break a few times because it couldn't take the stress of having hundreds of beads slid over it. A nice surprise is that the beads give the shawl some good weight, so it seems a bit less likely to slide off my shoulders.
Now I just have to wait for fall to really roll around for me to be able to use it!
It's almost become a cliche to say that we have so much information now that our biggest challenge is finding the relevant pieces, not making sense of them. And this is a fairly representative article cheerleading for the new technologies that will help with data search, focusing on data mining and the construction of a federated solution that can normalize across very different data sources with different base formats. But this is also a representative article in the misleading way that it blithely says what technologies will allow us to do, with no discussion of where we are now on these projects, what the hard problems remaining are, and realistically assessing how far away these solutions really are. As just a sampling of claims:
"By automatically classifying, summarizing, and discovering the "who," "what," "where" and "when" of each document, publishers, government organizations, and enterprises can do more than ever before -- on a massive scale."
"Visualizations allow users to quickly sift through and locate information and patterns in hierarchical, relational, tabular, or time-based data sets;"
"This new generation of solutions will need to go deeper than keyword search -- it will require a deep understanding of language, to lend structure to unstructured data for use in downstream analysis and assessment."
and the closing statement :"The world's data is at your fingertips - where it should be."
There is a single acknowledgment of fallability when it is mentioned that people can look at the output of an information extraction system to give feedback to improve the accuracy. But overall, this article reads as if all of these promises are a year or two away. Particularly when it comes to all of the claims about systems being able to handle widely diverse data sources in widely different structural formats, including free-form text, this is just flat wrong. On top of that, the ethical barriers to developing and deploying such solutions aren't mentioned at all.
All of the technologies mentioned do come out of the problems researchers want to solve, but the business community is not going to get the universal tools that are described here any time soon.