dubious.biz(ness)

I just upgraded to a new version of WP, and moved to a different box. I hope to make a few changes to the blog so that it can be used as both a subject oriented blog for “Going to College” and also be useful as a general blog. Of course, I don’t have a lot of time on my hands and right now I’m more concerned with getting the server up and running.

This box is a Mac Mini running OS X, which I have to say is pretty cool. I would pass along a few pieces of advice: unless you are cheap, don’t do as I did and go with the regular MacOS X, upgrade to OS X Server or to a Linux distro. There are just too many hoops and silly little things to jump through to make the Mac work.

Secondly, a lot of those hoops are related to Very Good Ideas that I really think that modern open source *nix OSes need to consider but don’t. In short: you can really tell that a lot of smart people had a good hand in designing a scalable and easy to understand architecture. I’ll elaborate on this more at a later date, but I think that it is truly a compliment that Mac OS X is so usable out of the box as a Unix server, yet incorporates an overall smart design. Linux and the related million open source applications often seem to fall into one of a few architectural traps: Reinventing the Wheel, Overdependency, Semantics, and Backward Compatibility to a Fault.

I’ll only elaborate on the last right now. To some extent, software must be backward compatible. It is necessary for a word processor to read files from its own previous versions. It is often necessary that an operating system be able to run programs written for the previous version. However, architecturally, you must move forward and you must do so in a one-way fashion. There is no architectural “going back.” Too often OSS projects get bogged down into previous versions because the lack the vision and management to recognized a played hand. Or, worse yet, they see every new problem as requiring a different sort of architecture and never build in the necessary scalability, or they build too much in and never finish. All these are mangement pitfalls.

However one of the more upsetting is the general lack of communication that OSS software has with vendors of software the essentially design the environment that OSS goes in. RedHat, Debian, etc are basically the people who turn a bunch of packages from a bunch of different developers all solving a small problem into a usable system, and it is on them that the burden must fall to make key decisions about the direction of the OS. Why are all configuration files in such a terrible state? Why are log facilities barely abstracted to the system level? Why are security and ACLs still in 1972? These are the sorts of hard decisions that are programmed into a piece of code that can be run but decisions that must be made by strong management of the system itself by wise managers who care about the ultimate usability of the system. Many people would claim that we are too far along the current path to change something as basic as the file system, but is this really any different than deciding to make a 32 bit protected OS, or a registry to store all settings, or any other the number of things that Microsoft has implemented in the past 10 years? More importantly, is it any different that the changes that Apple has made to OS X to make it more future compliant than backward compatible?

The other three I will save for later, but the next time you are sudoing everything on a single user system (that is those of you who haven’t ’sudo password root’ and started cheating), consider that while sudo is a pain in the ass on such a system, it is the right choice for not just “dummies” or typical mac users, but everyone, and it makes the system better for someone finally having the balls to turn off root.

I challenge RedHat and Mandrake and all the other would-be OS makers to do the same and do things for the good of their OS, even if they will make propellerheads gripe, make all the textbooks wrong, and make more work for themselves because they are ultimately creating a better system.

As I said I’ll elaborate more on the rest later, now I need to continue on setting up this box.

A little note for all those who may be confused about compscis. Apparently it is not pronounced as it would be logically expected, as in the first syllables of computer and science. It’s pronounced “comp-skee”, as if it were Russian. I have no idea why it’s pronounced that way, a suggestion was made that Russian things sound cool, which seems at least plausible.

More likely, though, compsci has fallened victim to the same culprit as the rest of the English language: damn pronunciation, grammar, and syntax, it just sounds better. Otherwise we’d be speaking German.

At any rate, I am a compsci, which seems to be akin to being undatable and of the highest order of nerditude, even in England (which is choc full o’ nerds), along with sounding cool.

I guess I have been neglecting the blog lately for a number of reasons. First of all, this is the long wait/work for me. I have to figure out where to get the money, prove I can get it, and get a reasonable grade in school.

So far, it’s been pretty apparent that its going to be a lot of work on my part. Suffice to say there is not a lot of hand-holding from the college. This may be one of the key differences. I think because of my age my parents mainly expect me to do the legwork, and the college basically expects that someone will do it without really caring who does it.

In other words, the burden of proof and ultimate responsibility lies on me to secure funding. This is a bit of a turn from if had gone to school in the US, and probably one of the most difficult parts that certainly wasn’t obvious when I started. However, I am fortunate enough to be in a country where I have a lot of finance options available to me (compared to, say, Africa or Asia). Of course, even loans and grants would typically be handled by the financial aid office, and there is no such office for international students, here or at Cambridge.

Certainly the amount of money available to me makes it possible, particularly in Stafford loans, however, do not expect anyone to help. I am trying to coordinate three agencies: the bank, the school, and the government, with very little assistance from the college or the British Council (the marketing arm of the British educational system :-) ). The British Council has been by far the most helpful in giving me pointers on how to handle the bank, but most banks so far have been baffled.

I believe I’ll probably end up at Bank One, which is nice because they have a local branch and they have branches in London (as Chase Manhattan), although they do not allow you to move money directly between the two. I will probably have to get a second offshore account at HSBC to do that.

Also, I sent off my “bursary” application (think school grant). This bursary is very little known, even to me. In fact, were it not for the fact that it basically allowed me to write a “blank check” of how much I would like, I probably would not actually be considering still going to school due to the difficulty in getting funding. I am applying for about 5000 pounds, which is about $9000 right now. We’ll have to see how much of that I will actually get, but from the looks of things, I may be able to get at least 80% of that, as I have spoken to another student who was able to fund about that amount and who is also a computer science student.

Unfortunately, my chosen college, while comfortable and hospitable, is not very wealthy. I would certainly advise students considering going to school at Cambridge from overseas to balance the likelihood of them being offered a place with the college’s ability to fund them directly. Trinity, for example, is well known as one of the richest colleges in Cambridge and has more than a little money to throw around as such, offered in the form of bursaries for a number of different criteria, and more importantly, they have a large discretionary fund that generally goes to people who are awarded Overseas Trust/Commonwealth Trust bursaries (like the one I am applying for). This means that your college may kick in 1500 pounds on top of the funding from the Trust. My college is not likely to be in a position to be able to do that. However that said, I think it is probably an excellent fit for me, and while it would have been great fun to stay in a castle for three years at Trinity, I don’t know that I would really fit in with a bunch of obnoxious 18 year old brits.

Anyway, time for compsci, I will try to update this when I am in a better position with the loans and what not.