The article poster should do some research into the newish computational calculi, which were designed to express these very ideas. Object calculus and relational calculus would be good places to start to get a semantic foundation.
The Evilll electrician of U-571 tries to signal the destroyer overhead, and someone finally kills the evilllll SS-Nazi Gestapo Sea Killer electrician! Yay! Onward with the gratuitous stereotyping of our former enemies!
He's the frickin' captain! He lied about being the electrician!
Sheesh, if the original reviewer is too thick even to pick that up, it's no mystery why it took him two years to write the review!
I'm in my first open-plan office in three years, after having a private office. I'm also dealing with a bunch of type-a consulting types, who need their questions answered NOW NOW NOW. For them, I wear open-air headphones that fit over my ears, so they think I can't hear them. This encourages them to use IM (still an interruption, but they don't type as much as they talk) or email. It also leaves me able to hear peripheral conversations, so I can take the headphones off and jump in when necessary.
The only time I listen to music is to drown out excessive noise or lyrics-laden music from elsewhere in the office. The CEOs think music == good working environment, and have an amp stack in the corner of the main office area. When the music comes on, it's time to go home, because no work is going to get done. If the volume isn't high (rare), I crank up repetitive techno as white noise, so my brain doesn't kick into language parser mode when it should be in code generation mode.
The other big big big issue is tools availability. I've been using VisualAge/Java for years, and am in the process of switching to IntelliJ IDEA. Decent method, field, argument, class, keyword, and template completion, and a decent classbrowser with hierarchy browsing support, are essential for coding without having to switch context to look through javadoc, when you just need to find out if the method is toUpper(), toUpperCase(), upper(), etc.
Drink a lot of water, diet coke, whatever, so you have to take a lot of pee breaks, hopefully every 20-30 minutes or so. Gives the hands a rest, and gives you that essential "time away from the keyboard" so you can refocus on the problem and see it from another point of view. Sometimes it's hard to think of the solution if you're coding it up.
Thanks to the warning in this article, I have taped over the activity lights on my wireless ether card. Why go to the trouble of securing my connection with WEP if I'm just going to let the data fly out into the atmosphere with only minor protection???
That's silly. I'm assuming you pay for the computer you use at work, the portion of the network bandwidth you use, your chair, your desk, your phone line, your cell phone, and your pager?
It's a business expense, and companies are used to paying it.
Maybe your company has fooled you into paying to do more work for the same price, but not everyone is that naive.;>
Do you still remember the days when you were in college, and $10k would pay your tuition and room-and-board for a year at a state school, AND keep you full of beer?:-)
...I'd have to pick Java if it was up to me just because of its sheer elegance.
While C# may seem less elegant to you, it's because you do not understand it. For the first stretch of time developing in any new language, most efforts tend to be inelegant attempts to map previous knowledge onto a new construct.
I'm sure you've heard "writing x in y", like "writing lisp in c". What you attempted (if you actually wrote any code in C#) was "writing java in c#".
I find the code produced with C# tends to be more elegant than much of the code written in Java, simply because of the syntactic sugar available with the language. The Java language is entirely too redundant for maximum productivity. Many of the great gains of C# are in reducing the redundancy of Java without reducing the expressibility.
This isn't a slam on Java, although I could make it one. Anything new should build on the gains made by something that came before. In fact, this is my biggest complaint with most programming languages -- they throw out gains made from purer environments such as Lisp and Smalltalk, in the name of general palatability at design time.
Is a a single pointer to an array of 10 characters, or is it an array of 10 pointers to character? Unless you know the precedence conventions in C, there is no way to find out.
[reeses@ernie reeses]$ cdecl
Type `help' or `?' for help
cdecl> explain char *a[10];
declare a as array 10 of pointer to char
cdecl>
"...I'm not sure this site could survive even a slight slashdotting, which is why I may not have found it in the archives." Well, there's only one way to find out. We'll try the "early morning" timeframe and see if it survives.
What an egregious lack of courtesy! Someone expresses suspicion that the site might not take the load, and you post the article anyway, under the disclaimer of being able to pull the article if you trash his server? What, you go over to help your mom get ready for thanksgiving and jump on all her furniture so you know that if you break it, you have time to take it into the garage before Fat Aunt Bertha shows up?
Punks. At least be apologetic about your complete lack of etiquette.
Time for a history lesson
on
Tiny Apps
·
· Score: 1
The sort of thing you're asking about has been available for a decade -- with NEXTSTEP. I believe OS/X inherited this, but I could be incorrect.
Some time ago, I was looking for a solution to a cross-platform KVM integration issue, tying some SGI gear and my PC to a single console.
My problem arose because the SGIs use a sync-on-green signal through 13w3 plugs, while PCs typically don't. I was lucky in that most SGI kit made in the mid-90s and after uses PS/2 mice and keyboards. I further complicated my situation by wanting to tie the sound from the different boxes together.
Eventually, I bought a Belkin OmniView SE 4-port PS2 KVM switch, and some adapters. To handle the sound, I bought a cheap Fostex digital mixer from musiciansfriend.com[1], wired it all up, and today, I am the proud owner of a monstrous rat's nest of cables and boxes that pipe sound from multiple machines into one set of speakers.
It would have been a hell of a lot easier and cheaper just to buy three sets of speakers. I'd be careful when evaluating a KVM+Sound switch, verifying that sound is mixed from all channels, as it is almost a necessity to be able to hear output from a non-focused computer.
As it is, I'm ecstatic about the Belkin gear. I have the OmniView SE here at home, and the OmniCube at work, and with decent cables, I can drive the monitors at 1900x1200 without any difficulty. I found this to be just about the most important piece of the kit -- with cheap cables (the kind you can buy for $10), the video starts bleeding and exhibiting static above VGA resolution. Don't waste your time with the cheap cables, spend the extra $10 or so and get the decent ones.
[1]: I am intentionally not making that a link because the bastards are almost as bad as x10 for daily emails once they have your address.
I completely agree. We had a horrendous problem with JSP proliferation, getting to the point where we had hundreds of JSP pages for a relatively simple application. Basically, as web developers learned some Java, they'd use it. In and of itself, that's not a big deal, but they really didn't know how to use it, and it became quite a nightmare. They'd take advantage of the servlet base, resubmitting temporary results to other JSPs, confusing and confounding the information flow.
We rearchitected the system around XML/XSL, assigning the webmonkeys (ahem) to the XSL layer to compose views based on a standard XML document structure. Client requests were dispatched to function handler objects by a mediating servlet, and then to the back-end business logic.
We lost a lot of flexibility by eschewing JSP, but that was part of the goal, which was the result of a lack of discipline on the part of the presentation developers. Taking away a lot of their options didn't solve all of the problems, and introduced some new ones[1], but it simplified QA remarkably and reduced my headaches.
[1]: We had to operate in lock-step. As presentation is sometimes much easier to develop than business logic, the XSL developers often had to wait, none-too-quietly, for adequate function handler and XML-generation code.
I had the advantage of living near an excellent audio store here in Seattle, of which I availed myself before buying my audio gear. They have an excellent page on exactly how you should structure your listening tests: you should listen to the tune.
(The rest of this comment is a small rant; feel free to ignore it.)
I ended up buying my integrated amplifier from NAD (which has the added benefit of a humorous name), CD player from Marantz, and speakers from Axiom. (You can't imagine how good these inexpensive speakers are.) I aimed for decent sound in my office without spending too much money. I'm not an audiophile, but I also know a particularly obvious imperfection will bother me. The system is pretty well-balanced for the price.
After replacing my crapomatic integrated bookshelf stereo in the office with decent components, the issues of MP3/OGG/etc. became irrelevent. They're all crap, and your only options are separating less stinky crap from really stinky crap, if you're listening to generated waveforms coming from your internal DAC. My SGIs, which spank any PC with an internal sound card for fidelity and noise level, don't even match up.
I really want to try one of the USB-connected external DACs, because I like the ability to manage my entire CD catalog from one place, without having to switch CDs. However, I couldn't continue kidding myself that the sound from the computers was the least bit faithful.
The main advantage is here: development is faster in a team where every programmer can use the language he/she likes the most.
Do you have figures for this, beyond,"I like Perl, and gosh darn it, I can code Perl faster than I can code Befunge, which I hate"?
I propose the exact opposite -- a team that standardises on a single "good enough" language is more productive. Not only can everyone read and edit everyone else's code, more advanced team members can mentor junior members, leading to a greater average individual ability. Being able to ask for help is also a huge advantage.
Given this, I would prefer to see a list of operating systems in which things were done RIGHT, but which are no longer in use or from which lessons are not being learned.
Multics, TOPS-10, and TOPS-20 come to mind.
An excellent sign that you've never even looked at Multics, and therefore probably have no idea what you're talking about.
Slashdot Mirror
The article poster should do some research into the newish computational calculi, which were designed to express these very ideas. Object calculus and relational calculus would be good places to start to get a semantic foundation.
Google will show you the way.
Umm, ready to kill an idiot for not decomposing a 200-line function into smaller functions?
There are statistical optima for function lengths with respect to defect levels, and 200 lines is not it for C++ code.
Anyone who hasn't read that book by now has no interest in producing decent code. :-)
He's the frickin' captain! He lied about being the electrician!
Sheesh, if the original reviewer is too thick even to pick that up, it's no mystery why it took him two years to write the review!
I know I'm volatile and prone to inflammability after downing a few measures of Lagavulin!
I'm in my first open-plan office in three years, after having a private office. I'm also dealing with a bunch of type-a consulting types, who need their questions answered NOW NOW NOW. For them, I wear open-air headphones that fit over my ears, so they think I can't hear them. This encourages them to use IM (still an interruption, but they don't type as much as they talk) or email. It also leaves me able to hear peripheral conversations, so I can take the headphones off and jump in when necessary.
The only time I listen to music is to drown out excessive noise or lyrics-laden music from elsewhere in the office. The CEOs think music == good working environment, and have an amp stack in the corner of the main office area. When the music comes on, it's time to go home, because no work is going to get done. If the volume isn't high (rare), I crank up repetitive techno as white noise, so my brain doesn't kick into language parser mode when it should be in code generation mode.
The other big big big issue is tools availability. I've been using VisualAge/Java for years, and am in the process of switching to IntelliJ IDEA. Decent method, field, argument, class, keyword, and template completion, and a decent classbrowser with hierarchy browsing support, are essential for coding without having to switch context to look through javadoc, when you just need to find out if the method is toUpper(), toUpperCase(), upper(), etc.
Drink a lot of water, diet coke, whatever, so you have to take a lot of pee breaks, hopefully every 20-30 minutes or so. Gives the hands a rest, and gives you that essential "time away from the keyboard" so you can refocus on the problem and see it from another point of view. Sometimes it's hard to think of the solution if you're coding it up.
Thanks to the warning in this article, I have taped over the activity lights on my wireless ether card. Why go to the trouble of securing my connection with WEP if I'm just going to let the data fly out into the atmosphere with only minor protection???
That's silly. I'm assuming you pay for the computer you use at work, the portion of the network bandwidth you use, your chair, your desk, your phone line, your cell phone, and your pager?
;>
It's a business expense, and companies are used to paying it.
Maybe your company has fooled you into paying to do more work for the same price, but not everyone is that naive.
The heads of babies, crushed under the bare feet of their parents, mortared together with feces and broken glass?
They'll never get it to play track 2 from Windowlicker. (Although, I do love Amazon's attempt.)
You're getting old.
:-)
Do you still remember the days when you were in college, and $10k would pay your tuition and room-and-board for a year at a state school, AND keep you full of beer?
But did it take you 750k lines to come to that conclusion? ;>
While C# may seem less elegant to you, it's because you do not understand it. For the first stretch of time developing in any new language, most efforts tend to be inelegant attempts to map previous knowledge onto a new construct.
I'm sure you've heard "writing x in y", like "writing lisp in c". What you attempted (if you actually wrote any code in C#) was "writing java in c#".
I find the code produced with C# tends to be more elegant than much of the code written in Java, simply because of the syntactic sugar available with the language. The Java language is entirely too redundant for maximum productivity. Many of the great gains of C# are in reducing the redundancy of Java without reducing the expressibility.
This isn't a slam on Java, although I could make it one. Anything new should build on the gains made by something that came before. In fact, this is my biggest complaint with most programming languages -- they throw out gains made from purer environments such as Lisp and Smalltalk, in the name of general palatability at design time.
But that would make this a rant.
We agree, then. :-)
Tools that make you stronger, faster, better aren't necessarily crutches. Think of it as a bionic leg, not a crutch. :-)
The less I have to remember about the rudiments, the more I can think about the application.
[reeses@ernie reeses]$ cdecl
Type `help' or `?' for help
cdecl> explain char *a[10];
declare a as array 10 of pointer to char
cdecl>
Learn your tools; they're there to help.
What an egregious lack of courtesy! Someone expresses suspicion that the site might not take the load, and you post the article anyway, under the disclaimer of being able to pull the article if you trash his server? What, you go over to help your mom get ready for thanksgiving and jump on all her furniture so you know that if you break it, you have time to take it into the garage before Fat Aunt Bertha shows up?
Punks. At least be apologetic about your complete lack of etiquette.
The sort of thing you're asking about has been available for a decade -- with NEXTSTEP. I believe OS/X inherited this, but I could be incorrect.
Read this and this for a start.
The easiest and best way to emulate vi on Emacs is to evaluate the expression "(use-global-map (make-sparse-keymap))". Go ahead, put it in your .emacs.
Just like vi, it will beep every time you press a key, and you won't be able to exit.
LOL. "Sure, I've done software architecture!" == "I took an OO architecture class, and designed a fifteen-class system!"
Some time ago, I was looking for a solution to a cross-platform KVM integration issue, tying some SGI gear and my PC to a single console.
My problem arose because the SGIs use a sync-on-green signal through 13w3 plugs, while PCs typically don't. I was lucky in that most SGI kit made in the mid-90s and after uses PS/2 mice and keyboards. I further complicated my situation by wanting to tie the sound from the different boxes together.
Eventually, I bought a Belkin OmniView SE 4-port PS2 KVM switch, and some adapters. To handle the sound, I bought a cheap Fostex digital mixer from musiciansfriend.com[1], wired it all up, and today, I am the proud owner of a monstrous rat's nest of cables and boxes that pipe sound from multiple machines into one set of speakers.
It would have been a hell of a lot easier and cheaper just to buy three sets of speakers. I'd be careful when evaluating a KVM+Sound switch, verifying that sound is mixed from all channels, as it is almost a necessity to be able to hear output from a non-focused computer.
As it is, I'm ecstatic about the Belkin gear. I have the OmniView SE here at home, and the OmniCube at work, and with decent cables, I can drive the monitors at 1900x1200 without any difficulty. I found this to be just about the most important piece of the kit -- with cheap cables (the kind you can buy for $10), the video starts bleeding and exhibiting static above VGA resolution. Don't waste your time with the cheap cables, spend the extra $10 or so and get the decent ones.
[1]: I am intentionally not making that a link because the bastards are almost as bad as x10 for daily emails once they have your address.
I completely agree. We had a horrendous problem with JSP proliferation, getting to the point where we had hundreds of JSP pages for a relatively simple application. Basically, as web developers learned some Java, they'd use it. In and of itself, that's not a big deal, but they really didn't know how to use it, and it became quite a nightmare. They'd take advantage of the servlet base, resubmitting temporary results to other JSPs, confusing and confounding the information flow.
We rearchitected the system around XML/XSL, assigning the webmonkeys (ahem) to the XSL layer to compose views based on a standard XML document structure. Client requests were dispatched to function handler objects by a mediating servlet, and then to the back-end business logic.
We lost a lot of flexibility by eschewing JSP, but that was part of the goal, which was the result of a lack of discipline on the part of the presentation developers. Taking away a lot of their options didn't solve all of the problems, and introduced some new ones[1], but it simplified QA remarkably and reduced my headaches.
[1]: We had to operate in lock-step. As presentation is sometimes much easier to develop than business logic, the XSL developers often had to wait, none-too-quietly, for adequate function handler and XML-generation code.
I had the advantage of living near an excellent audio store here in Seattle, of which I availed myself before buying my audio gear. They have an excellent page on exactly how you should structure your listening tests: you should listen to the tune.
(The rest of this comment is a small rant; feel free to ignore it.)
I ended up buying my integrated amplifier from NAD (which has the added benefit of a humorous name), CD player from Marantz, and speakers from Axiom. (You can't imagine how good these inexpensive speakers are.) I aimed for decent sound in my office without spending too much money. I'm not an audiophile, but I also know a particularly obvious imperfection will bother me. The system is pretty well-balanced for the price.
After replacing my crapomatic integrated bookshelf stereo in the office with decent components, the issues of MP3/OGG/etc. became irrelevent. They're all crap, and your only options are separating less stinky crap from really stinky crap, if you're listening to generated waveforms coming from your internal DAC. My SGIs, which spank any PC with an internal sound card for fidelity and noise level, don't even match up.
I really want to try one of the USB-connected external DACs, because I like the ability to manage my entire CD catalog from one place, without having to switch CDs. However, I couldn't continue kidding myself that the sound from the computers was the least bit faithful.
Do you have figures for this, beyond,"I like Perl, and gosh darn it, I can code Perl faster than I can code Befunge, which I hate"?
I propose the exact opposite -- a team that standardises on a single "good enough" language is more productive. Not only can everyone read and edit everyone else's code, more advanced team members can mentor junior members, leading to a greater average individual ability. Being able to ask for help is also a huge advantage.
An excellent sign that you've never even looked at Multics, and therefore probably have no idea what you're talking about.