Slashdot Mirror


User: judd

judd's activity in the archive.

Stories
0
Comments
130
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 130

  1. Complete agreement - critique for free on New Essay about Hacking · · Score: 1

    (people who dislike pedantry should skip the following comment)

    What's wrong with the spelling, grammar or punctuation of "Taylorist principles"? It parses correctly, and "Taylorist" is an acceptable neologism for "of or pertaining to Taylor".

    And "avant-garde" is a noun in English as well as an adjective. Possibly the author may have meant "vanguard", but that's a malapropism, not a spelling, punctuation or grammar error.

    With you on "bourgeoisie society" however.

  2. Web site needs vs small business needs on What Database is the best for a Web Site/Small Business? · · Score: 1

    The original Q was "what's best for a website or small business?". I don't think these share the same needs.

    When you use a DB to back a website, speed is critical. The time to establish a connection to the DB becomes very important too (or the ability to maintain a persistent one and recycle it). On these grounds, MySQL/mod_perl all the way.

    It takes forever to get a connection to Oracle.

    (incidentally, has anyone with an advance copy of Oracle/Linux got Oraperl or DBI::DBD working?)

    However, the lack of subselects and union in MySQL is a pain in the arse. Yes, you can code around it, but this is a nuisance, and makes for messy code, since some of your query is in the SQL, and some in whatever procedural language you're wrapping it in.

    In a business setting you're more likely to miss these features.

    Another thing to consider is whether you'll be writing back a lot (or at all). Often for a website the DB is essentially read-only, in which case lack of transaction, commit or rollback is forgivable. MySQL wins again. OTOH, business apps are not like this.

    One possible solution might even be to use MySQL to back a website while using a more sophisticated DB in the business, and dumping into MySQL overnight.

    Moral: it all depends. Create your own checklist of needs and then do your own comparison.

  3. A classic example of poor web maintenance, too on Classic Computer Science Papers · · Score: 1

    The sad thing is there don't seem to have been any updates since mid-1996. Don't you hate continuing series that don't?

    I guess the ACM don't care about their website that much. (And they're using the CERN webserver still!)

  4. Where's the proof M$ is reverting to subversion? on Feature:The Two Towers · · Score: 1

    One word.

    FrontPage.

    I spent an hour yesterday with a customer who had just bought FrontPage 98, which relies on having an IIS server to talk to and MS' own proprietary implementation of SSI.

    (No, I don't want to install the FrontPage extensions to Apache)

    The customer believes it is our fault, of course.

  5. Beating and winning are improper goals on Feature:The Two Towers · · Score: 1

    I'm glad this article was written.

    It's been bugging me for some time to read things like "Linux must do X to beat Y".

    I don't think there is a "Linux must". There is just a bunch of people writing code that pleases them, meeting whatever goals that they as people have.

    Once we stop focussing on writing code that meets personal needs, we get locked into stupid dicksize wars. It's OK if beating MS is a happy side-effect, but making it the goal is really dangerous. Microsoft's goal is to beat everyone else, and look at the code they produce...

    It has already been observed on ./ a couple of days ago that the most successful free software is written to meet a pressing need on the part of the author, rather than to tick off a feature in comparison with a commercial vendor's offering.

    We should stop wanting to beat Microsoft (or Oracle, or Sun, or whomever) and begin wanting to write great free code (or if you don't code, help people who do.) All else is distraction, and dangerous.