279 karma · joined February 22, 2007
My first macro in Arc' would be "arcp" a macro which took Arc code and translated it to Arc' code. Then you just wrap every file in:
(arcp ;;file contents )
So, based on that, Arc' has an extra macro compared to Arc and extra entries in each file, and thus isn't as succinct as Arc and thus isn't as powerful as Arc -- if we're going to be philosophical about it.
I had just one day earlier spent an hour on the phone with a customer because a program I wrote actually uses the correct ProgramData location, but the program hadn't been added to the User Account Control (UAC) yet so Vista was preventing it from writing some files. Once added, it worked perfectly. Who wants to deal with that crap?
Long story short: Ubuntu costs me $0, it works with my dual monitors, it was easy to install, the second hard drive got mounted easily, and I'm switching to web programming. I'd also like to say I am in love with Ubuntu. It's like rekindling a flame with a previous lover (Linux).
I do think that you should know probably PHP (for consulting most web projects are in this or are extensions of projects written in it) and one of either Ruby or Python.
There are opportunities out there, just not nearly as many. Also keep in mind as a recent graduate you are competing with all the people who never made the web transition because they already had 3-5 years experience on the desktop that now have 10-12 years experience there.
char* p
instead of
char p
That way there was no clash with when you are dereferencing p -
p
Admittedly this approach may not help entirely with more complex examples, like a pointer to a function that returns a point to a function that takes two pointers to void...or something. Indeed having languages that create less accidental complexity is a huge win.
First, for his evidence he is using a study from the Project for Excellence in Journalism. Do you think the study might be biased? The crowd mentality threatens the entire profession of Journalism (I'm not saying it's going away, but they are less and less in control of what people read hear and see).
Second, he argument assumes the dividing line is "hard news" vs. "soft news" when it is really "I care" or "I don't." The problem is that nobody cares what's on the news. If we are to believe the study, the news is more consistent in what it shows. That means they are showing us the same thing over and over again. People care about things that are relevant to them or that provide some form of entertainment. Surprisingly, the mainstream news is just picking up on the fact that their product at the very best under-serves the needs of their market.
Third, the article paints the mainstream media as "informed" and everyone else as "uninformed." No bias there. Even if we assume they are right, it just points out that they aren't providing their customers with the information they want. If Google can target ads to me based on search, how come news can't be tailored to my desires.
Finally, the crowd based sites know the mainstream news exists. They don't have to cover the Iraq war for the 10,000,000,000th time. That's what the news is for. If you wanted to see information about Iraq you'd go to a mainstream news distribution channel. Haven't they considered that maybe there is no reason to rehash those stories already beaten to death in the mainstream media?
1. Big-Design-Up-Front
2. Have a testing department
3. Use Source Control and Bug Tracking
4. Fix all instances of systematic errors and fix the process by which those errors were injected into the program.
Most of those sound familiar:
http://joelonsoftware.com/articles/fog0000000043.html
I admit that this article isn't written as a technical instruction sheet for writing perfect software, but they could at least come up with something we haven't heard of before. You mean NASA actually has the funds to do a full BDUF and to have a testing department? No kidding.
"Some of my banking customers have infrastructures that are bigger than the public Internet in terms of the storage they consume, given all the data and equipment they have to support."
Somehow, I doubt that anyone has a system larger than the public internet particularly as measured by storage. Can you imagine trying to sell that much hard drive space to someone? Yeah, you need more space than the Internet! I bet they aren't even using as much space as YouTube. I tuned out after reading this comment in the article.
Ever since I read PG's article about PR I've been leery of articles that stink of PR. This article reeks of it and then he goes and makes an incredible claim with no backup. There are hundreds of other articles I could be reading, I'm done with yours.
Would you rather start programming on the surface (not knowing what is happening underneath) or would you rather start by learning the hard stuff underneath? I, personally, would rather know what is happening underneath. Is there an argument for learning the more surface stuff first? Maybe I'm off nuts and have a slanted view because I learned Java, Scheme and then C.
I also try to write code so small and simple that it approaches the complexity of "Hello World." That way, it's easy to find the bugs.
If you want to get stuff done quickly, learn Python or Ruby.
Be is harder IMO. Do you want to be a husband? father? grand father (you need help on this one obvious)? Do you want to a champion whatever? These are the questions I would ask myself. With the answers, you can start taking concrete steps.
Also, are we counting things like macros in this estimation? I mean, a single line macro could yield hundreds of lines of code once through the compiler.