 Bird, bird, bird; bird is the word.
I am taking today off to eat a shitload of turkey and drink gin until football becomes entertaining.
Normal (that is to say, semirandom) updates resume tomorrow.
 Cory Doctorow has an interesting article he posted while cruising about high above the earth in his blogosphere about an immune system for the internets. He brings up some curious points about how difficult it is to unfuck the immune system once it thinks something fishy is going on, and then goes on to compare the immune system's issues with how dangerously hard it is to get a computer system to acknowledge that you are in fact harmless once it identifies you as a threat. This metaphor of the internet as a body with an immune system made me think of the other implications contained therein.
Doctorow suggests that perhaps something should be done to automate the unfucking of these systems. For example, he recalls staying at a hotel whose network decided his laptop's frequent pings towards other hosts on the network were the start of some malicious port-scanning, when in fact it was just a game he was playing on his laptop's attempts to find other hosts to play with. If only he could hit a reset button to restore internet access to his hotel room as easily as he tripped the "DANGER HAXX!" tripwire!
Well, why can't he? Just like we have sophisticated immune systems built over time, we could rig networks with sophisticated immune systems built from experience. Why not have multiple layers of defense, with incremental response? Instead of just hosing anything looking for hosts, the hotel might consider this a step to future hosing. First, is the offender checking many ports all over? Yes. Branch down MULTPORTS path of the decision tree. Instead of cutting access here, seek to understand what the program is doing. Check the packet payload. Hmm, its structure looks similar to another branch of the tree, GAMEPACKET. Use Bayesian filtering to decide that it either is or isn't a game packet. If it is, let it go about its merry way. If it isn't, seek other possible intents that match the packet. If none are found, kill the program. If it is legit, the man in the hotel room will call up and complain, and you can add another layer to your filter.
This sort of ad-hoc system for determining intent would interesting to implement. Perhaps it would find too many false positives to be useful, but it is interesting to think of the ups and downs of an internet immune system, and why it is worth attempting to fix it or not.
 A VC blog I chanced upon a while ago contained this lovely epithet, which was a source of some consternation for one of the bloggers who felt that this phrase (used as an appraisal of the VC scene in NY) was too harsh.
And yet I can't help but think that it may be an apt description for the startup scene in Boston. My comrade Mr. John Watson wrote me an email yesterday with this content:
Were I in 'Frisco, I'd attend stupid shit like this: http://www.facebook.com/event.php?eid=5907866589&ref=mf People say that Boston is a startup & technology hub. What am I missing? Is there initiative that needs taking?
After some thought, we decided we were right. There was initiative to be taken. And who better than some idiots who want to start a startup?
So I put it you, loyal MOCDL readers (all 2 of you). What should we Bostonians do? Who would be fun to invite to speak? Or would we be better off with just a meet and greet at a cool location?
 During my final few months at UNH, I have been taking a class by the name of "History of the Modern World" which UNH, in its infinite wisdom, decreed I should take to make sure I don't come out of college completely ignorant of the past. On the surface, it is an entry-level history major class; however, lurking beneath the surface is its true function, at least in my eyes: showcase of the stupidity of UNH students.
Now let's get some stuff clear. I'm going to berate two groups of people here: UNH students and history majors. I love history: it is one of my favorite things to read about. I hate history majors: they are lumped in with pysch majors and communications majors in a group I like to call "My Future Cashiers". UNH is a decent enough state school, which I ended up at after a year at WPI completely wiped out my savings.* I like UNH: clean, quiet campus. Wonderful professors (in CS, anyway). And above all: cheap.
However, because it is a state school, it often reminds me more of a high school than a university. The amount of popped collars exceeds the physical number of people on campus.** The goal of most of these kids seems to be nothing more than to get laid, with alcohol a close second. These are fine pursuits to be sure, but when your hung over ass almost throws up on me while I try to dodge your stumbling attempt to cross the street at 9 am on a Wednesday, I may grow a little irked at your dedication to your craft.
However, I have mostly high-level CS courses, notably devoid of this stereotype. The only gen ed I take this semester is the aforementioned history class, which is full to the brim of people that remind me of high school, and why it fucking sucked. One kid in particular's insistence that Communism worked perfectly really took the life out of me today. It seems like everyone has this kid in history class at some time in their life: fat, strident, ill-washed, and smugly sure of his superiority over the rest of these sheep who think capitalism is the bee's knees. His stained Che shirt, which he wears without a trace of irony (slightly distorted by his massive gut), is as much a symbol of his stupidity as it is any resistance against The Man.
Now, I remember in WPI this kind of kid existed as well. For some reason, I took a Shakespeare class there, and a version of this blob manifested himself in that classroom as well, often loudly opining that we were idiots and dolts beyond measure for thinking that Shakespeare existed. He was OBVIOUSLY a group of different authors working as a collective!***
The difference at WPI is that the professor fucking called him out, asking for evidence that that was true, and citing his own opinion that Shakespeare was quite real with well-reasoned arguments. Clue entire class chuckling at cookie-cutter nonconformist getting owned.
But at UNH, this never happens. The professor thanks him for his valuable contribution, and I groan and slouch down in my chair. Is this worth it? What the hell am I getting out of this exchange? Truth be told, I spend the vast majority of time in that class thinking about startups. They hold more interest for me these days than anything UNH has to offer.
* Thank you, three consecutive tuition hikes over the course of one year! You made my already 40K a year education completely unattainable!
** Seen crossing commuter parking lot: A kid with no less than three popped collars: two shirts, one fashionably rumpled black blazer. And a tan no one in New England could ever have.
*** Cue lecture on the uses of collectivism, with frequent interjections about how much he wishes he could suck Lenin's preserved dick. Or maybe I'm just reading between the lines there.
I kid. Some of my best friends are Germans!*
Weitz is mostly cool because of cl-who and hunchentoot. As I have let slip before whilst rambling about the internets, my friend John Watson and I are hard at work on a fun project involving Common Lisp. While he fiddles with memcached, I am working on the beginnings of our site, which we plan to build using Common Lisp to dynamically generate HTML and skin it using CSS to make it look all pretty and WEB 2.0**.
This bold plan would be a hell of a lot harder if not for Weitz, and his wonderful libraries hunchentoot and cl-who. Weitz, self-effacing devil that he is, is quick to admit that many other Lisp markup languages exist. And they do. But few work so seamlessly with his major accomplishment, hunchentoot!
hunchentoot is a hilariously named*** Common Lisp webserver. Currently running on its own right now, it is serving me the very first test pages for our site. I am extraordinarily pleased with it so far, and cannot recommend it highly enough.
This is a pretty short update for today (lacking even a lolcat!) because I have an electrical engineering exam to study for and some work to do for my day job. I may throw some more up later tonight, but most likely I will just collapse somewhere. The lazy is strong with this one.
* Patently untrue. I only know one, whom I met twice. He dates my girlfriend's best friend. Regardless, he is awesome, and has raised my opinion of Germans in general.
** Rounded corners. Lots of rounded corners.
*** I am forbidden from talking about it in public, due to the odd glances I have started to garner: "Something smells fishy with hunchentoot!"
 I often wonder about what my life would be like if I continued to work as a developer for large companies for the rest of my life. I can see two possible paths:
1) Stay Programmer.
2) Become Manager.
This is not a new dilemma. I'm sure people have faced it before with jobs in various fields other than software development. But it takes on a vital air when discussed among programmers and/or managers because the two are so different, and yet often one is promoted into the other.
Programmers* are very similar animals. They are all usually less than outgoing, interested in solving curious problems, and inclined to focus on one thing to the point where everything else becomes a dull background roar.
Managers are outgoing people. They have to be if they are any good at their craft, which is the sole act of dealing with and coordinating people. They enjoy nothing more than working with people, talking to people, and thinking about what to do with people and their time. They focus on a wide variety of tasks and they need to be able to context switch quickly.
Now you may think that I am about to start slagging managers as useless and talentless individuals. And some are. But a good manager is an extremely useful thing. He solves logistical problems, and aids the programmers with whatever is holding them up. An enabler who helps people get shit done, with a minimum of intrusion otherwise.
Nine times out of ten, programmers make rotten managers. They hate talking to people, they hate meetings longer than 10 minutes, and most of all, they hate to stop programming, which is essentially what you must do when you become a manager. Once in a while you will see an excellent manager promoted from a programmer who truly understands his employees, and is an absolute joy to work for with a laissez-faire management style and a no-bullshit attitude. These are gems that must be coveted beyond imagining: if you work for a manager like this, keep working there. And make sure he doesn't get fired!
This is mostly just an unstructured rant about how annoyed I am at the preponderance of shitty managers in software development today. And it is a difficult problem to fix: while most promoted from programmers suck, at least they are in touch with the problems programmers have. Nothing is worse than some business-school idiot with no programming experience being tapped to lead a dev team.
So what is the answer? I would begin by offering MAJOR salary increases for programmers becoming managers, and still allow them to program as they see fit, and delegate some managerial duties to an assistant as possible. This will keep them focused, interested, and ready to lead without boring them. I like this model, and if I ever have to become a manager (it is sometimes unavoidable if you wish to keep working at a particular company), it will definitely be my goal to set up my reign in this manner.
* When I use the word programmer, I mean a real programmer. Not someone who is doing it because they thought a CS degree was a ticket to instant free wealth. Someone who would be programming even if it were a completely unpopular social stigma akin to having herpes, and if it paid nothing at all, and in fact cost one money. The people who understand and love their profession, which is writing software. My God, I have a chip on my shoulder the size of Mt. Everest, don't I?
 Freddie Mercury said it best: spending all your time and money to keep yourself alive.
I would like to go on record saying that I wholly endorse the trend of super-bootstrapped startups such as 8coupons, which is featured in that article. It pleases me beyond measure to see people going to such lengths to pursue their startup goals. Quite frankly it makes me feel extremely lazy and underachieving, picturing these people cranking out code with steady jobs while I work piecemeal at best, with only a part-time programming job and some undergrad classes to distract me. I should seriously take a page out of their book and man up, crank out some code, and get shit done.
In fact, I think I will! Screw this blog, I'm going to work!
...right after I go out for some drinks at Beehive and hang out with my girlfriend. Having a girlfriend in Boston is dangerous to your work ethic!
 One of the major dilemmas facing would-be entrepreneurs (at least judging by this thread at Hacker News) is whether or not to seek a co-founder once you have a viable idea.
Opinions differ pretty strongly about this. Paul Graham has espoused the feeling that startups with more than one founder are more likely to succeed. He goes on to suggest that two or three founders is the optimal amount. Many people seem to think that he is the king of all things startup, and so these same people think that his way is the only way.
Which leads to posts like this one. A bright guy is hesitant to start working on what is probably a good idea, all because he lacks a co-founder! This is the danger of relying too heavily on one theory of startups. This guy should probably just go ahead and start his idea, rather than look for a complete stranger to entrust half his company to.
On the other hand, there are also people like this. After reading that long-ass rant, I'd be disinclined to work with him as well. He seems to have a pretty high opinion of himself, and seems convinced of the brilliance of his idea to the point where he is unlikely to accept any criticism.
So what is the answer? You and your co-founder don't want to end up like Mr. White and Mr. Pink up there, at each other's throats with .45s. But if you go it alone, it is definitely harder.
As I posted in the thread, I think there is a middle ground. One should neither seek out a co-founder actively or ignore all potential co-founders. Just look amongst the people you have known for a few years or more, and decide if any of them share similar goals. If not, expand your horizons and look to meet some like-minded people -- not for co-founders, but to meet new friends. Among these, you may find a co-founder or you may not. Just continue to expand your network of friends, all the while working on your idea.
Just my two cents. Slow news day, and I actually had my coffee well in advance of posting today, so this is a much calmer entry.
 Time for a departure from my normally scheduled crap concerning lame-ass web startups* and stuff I like on the internet. This post is about COFFEH!
This post is also going to be profane. Watch those sensitive ears, ladies.
Coffee, as a rule, is delicious. I don't care how you imbibe it, be it lattes, cappuccinos, espressos, regular fucking coffee, whatever. It all tastes good and it all has a fucking stranglehold on my brain. It has gotten to the point now where a day without caffeine input (preferably coffee) leaves me incredibly irritable and with a skullsplitting headache.
As addictions go, this is a pretty tame one. I'm not sucking dick for coffee in alleys, or mugging people from the Midwest on vacation for coffee money. I should be glad that I am addicted to such a readily available substance. However, I am dismayed, probably just like any cocaine/heroin/janjaweed addict, by the vastly varying quality of my substance of choice.
Point in case: I am drinking a Dunkin' Donuts** coffee right now. It tastes like pee. But it is still much, much better than some coffee you can get. While it dismays me that this coffee is the best available to me on my way to UNH, it also makes me glad I am not drinking coffee from Cumby's, a disgusting gas station chain whose radio ads constantly shrill how awesome and trendy they are for serving Bengal Trader coffee ("ANY SIZE JUST 99 CENTS! YOU BETTER GET TO CUMBY'S!"). After some thought (not a lot, though), I decided to rank my coffee options:
1) Taste The Golden Spray
This is the highest ranking, reserved for the best indie coffee shops and other fine places where you can get some exotic shit to placate the monkey on your back. Also Illy, which I consider to be some of the tastiest mass produced coffee, unless you can get some Zabar's. The finest uncut Colombian.
2) I'd Buy THAT For A Dollar
This is my ranking for the best chains. I'd say stuff like Starbuck's and Seattle's Best rank in here somewhere. China white: it's been stepped on a little, but it's better than what everyone else is getting.
3) Mean Bean Machine
Dunkin' Donuts and all their ilk. Semi-decent coffee if you are in a hurry and don't need anything but a fix. Compare to Mexican Brown: tastes a little weird, still gets you where you need to go.
4) Cleveland Steamer In A Paper Cup
Gas station coffee. Any kind. It's all gross. Green Mountain is the best of this group, but still shitty. Badly stepped on Mexican Brown. Cut with Ajax.
5) SANKA TIME
Any decaf, anywhere. Hardware store complimentary coffee. ANY complimentary coffee. You're bootin' black tar heroin now.
And that's all I got today. Maybe tomorrow I will make fun of how useless Open Social is so far. Or maybe write about my amigo John's blog, which is so far even more content-free than my own, which is a feat indeed.
* Seriously, WTF. Like Amazon is impossible to use or some shit. Also their site design looks like one of those linkjack shitholes. As Ted Dziuba has been known to say, WARP FACTOR FAIL.
** Those of you from somewhere other than New England may not know what Dunkin Donuts is. I don't care.
 Or, How I Learned To Stop Worrying And Love Nick Negroponte.
Prepare to meet the best way to drop $400 dollars this holiday season.
Whilst in Miracle of Science, I struck up a conversation with a guy who happened to do some scut work on this project. I hesitate to let any of it loose, because it probably isn't for the public, but I learned some amazing stuff about this sweet machine. But let me just say, this is going to be an amazing piece of hardware and software. Buy it for some lucky kid elsewhere right now, or grab one for yourself as well on the twelfth!
|
RSS Feed