i enjoy working on projects where the interaction i have with people demonstrates to me that they care about what i’m doing.

i’ve happily worked on measure’s databases for so long because i usually get instant feedback from my boss and co-workers about what works and what doesn’t. though that’s entirely satisfying, i want to work on some new projects, several of my own design, which lack that feedback loop to urge me on.

i’ve heard and thus pondered on occasion about extreme programming (XP), a funny name for a practice that boils down to writing code in pairs: one person drives, types, and programs while the other analyzes, revises, and navigates.

always one step ahead, wired published an article this month entitled, “The New X-Men,” about two pairs of programmers at hp who love XP, as well as some others who don’t. it provides a readable introduction, and seems to capture what i feel i’m most missing in my present work.

of the twelve “rules” of extreme programming (from the wired article), i’ve exerpted the six most meaningful to me:

2. Small Releases Put a simple system into production quickly, then release new versions on a short cycle.
3. Metaphor Create an analogy that expresses how the parts of the new system work.
4. Simple design Design simply, and remove complexity at every stage.
6. Refactoring Edit the code to simplify, add flexibility, or remove redundancy.
7. Pair Programming Write all code with two programmers at one machine.
12. Coding Standards Use agreed-upon styles and nomenclature to promote easy understanding of what the code does.

i agree that it seems faddish taken altogether and not appropriate for all people or projects, but right now i’d die for that kind of interactivity so that i might sustain my grand web database/intranet/digital library plans. plus i almost have the power to hire someone into that role.

know anyone who’s interested?

friday at the mission is a defacto half day, so we came back to the hotel just after noon, and sat out by the pool reading and watching the ocean.

which makes me wonder how far out can i actually see, how far away is that boat on the very edge of the horizon?

with a little help from the pythagorean theorem and the interweb, i figured that if i was standing on the shoreline, i’d probably be able to see about 3 miles. hotel sea cliff just happens to be perched on a cliff (on the sea), so i can’t exactly stand at the shoreline, but assuming the cliff is about 40 feet off the water, i can probably see over 7 miles to the horizon.

i even made a cute little diagram:

solving for b:
a² + b² = (a + c)²
a² + b² = a² + 2ac + c²
b² = 2ac + c²
b= √(2ac + c²)

oh, and it helps to know there are 5280 feet in a mile

i use pine to read and compose email. many people call this “checking your email with telnet” as we used to telnet to isis.unc.edu in order to check our email on the library computers intended primarily for web-based research. actually telnet is just the protocol, analogous with the “http” of web browsing, but i suppose it does convey an air of 1970’s-era character-based displays.

normally i use securecrt to connect to ISIS to check my email with pine. this is ok on campus, with 100MBit/s connections, but off-campus, even with broadband, the character delay can be bothersome. (telnet and ssh work by sending each character you type as a seperate message to the server, which sends a message back saying “yep, i see that you just sent me an ‘s’–thanks!”. only once that acknowledgement returns does the character appear on your screen. if that message has to travel half way around the world or just across town but through a beleaguered server, you may type a few characters and not seem them appear until several seconds later. delayed feedback like this makes it very difficult to write an email.)

then there is file attachment, the killer piggy-back app of email, which is also difficult with pine because the files you want to attach must exist on the server (requiring a painfully cumbersome ftp or afs upload before anything can be attached).

i’ve often thought that i should create a client version of pine, until the other day i wondered whether the pine people at the university of washington might have already created one for me. low and behold, they have, it’s called PC-Pine, and it looks and smells just like everything i love about pine (ctrl+x ctrl+c, etc) while acting just like an outlook, mozilla mail, or mulberry.

the added bonus is how it takes advantage of the windows gui environment. you can use the mouse to scroll through messages, you can click on URLs, you can attach files from your local computer–and you can still use the keyboard for 100% of the program’s beloved functionality. (just make sure that when/if you’re installing it, you choose to store your profile locally.)

pine’s got a bazillion configuration options, and I’m still not sure how to send my password securely (using ssh or kerberos), or configure it for UNC’s ldap server, but if i happen upon anything interesting, expect to hear about it here.

i get about 10-15 spam emails a day. i delete them as soon as i see them. actually i bounce them to spam@unc.edu so hopefully someone at unc can contact the offending email servers. chances are they probably just delete what i send them. in any case it takes about a second to do, and it doesn’t bother me much.

but a lot of people (other than myself) are hyperaggravated about spam. from textism i found out about this interesting new spam blocking service called knowspam. if you read the details, you find that knowspam allows emails to reach you only if they meet any one of the following three criteria:

1. The sender is in your knowspam.net address book
2. The sender is in knowspam’s database of known humans
3. The sender correctly answers a question sent as a link by knowspam, basically a “look at this image and tell us what numbers you see” scheme (similarly employed by ebay and yahoo to prevent automated fraud).

Here’s the catch: “Once the sender answers that question, the sender may contact any other knowspam.net user because they have been added to the shared, verified-humans database.”

Is there not a hole there? All a would-be spammer has to do is get his/her email address into the “database of known humans” (by answering the knowspam question manually after sending a single email) and then they can get around knowspam’s entire contrivance.

Update: posed the question above to the folks at knowspam, and Thomas Burns provided the following: “responding to the challenge only lets you send 50 emails to people you do not know. Then you have to respond again. 50 emails is useless to a spammer.”

so if you get listed as a “known human”, you can take some comfort in the fact that you’ll only be listed for a short while, whereupon you shall have to prove your human-ness every 50 emails, ad infinitum.

i recently tried mozilla 1.4 rc1 and rc2. unfortunately it crashed every time i opened it (apparently i’m not alone). so i switched to mozilla firebird 0.6 (mozilla’s new browser-only product), but i had problems with the personal toolbar and missed the progress icon in the upper-right. it loads really fast, but my home page (google) seemed to take several seconds to load/render the first time. this first-time delay plagued a bunch of websites. so i’m back to trusty mozilla 1.3.1 and it makes me so happy to see my personal toolbar bookmarks and folder, and my calming pinball theme.

have i also mentioned that gestures rocks? it’s almost unbearable using IE once you’ve gotten accustomed to using mouse gestures. (for the uninitiated: imagine going back to a mouse without a scrollwheel). my fav recently discovered gestures preference: the ability to configure the “mouse trails” so you can see the gesture. too cool!