I don't have doc to back it up, but as far as Ive read, with the write balancing features of SSDs, they can last a sick amount of time before the amount of rewrites finish them off. Also, they have one big big big advantage over HDDs: you know in advance that they are going to fail, quite reliably. So sure, it MAY (not all that true depending on usage anymore) fail faster than an HDD, but it won't fail by surprise on a saturday morning.
Its still fairly different however. Declaring whats IN the food is a big step down from changing the food itself. I guess the equivalent would be to label software with the available accessibility feature and accessibility level of the software itself.
Having all that info on the box of packaged software (or let say, on an about page in a web site), that is, much more info than we have now, letting disabled people make better choices about the products they use, and allowing search engines to better categorise web sites, would go a long way (let say, if it was enforced in a law that the declaration page/document has to conform with the web site itself, even if all it says is "We do not have any accessibility features"), I don't think anyone would complain. (And it would help the non-disabled).
I can make food that will be lethal to a group of people (peanuts!), I just have to say it is. I still can use peanuts/sugar/milk/whatever as much as I want, and don't have to make the slighest effort to make my food more "minority friendly". I just have to be honest about it.
Section 508 is fairly all or nothing though. If someone actually cared to be accessible for their customer base, they'd probably look more at actual accessibility recommendations, or the W3C accessibility specs. If you want Section 508, its probably to fullfil a legal requirements, you have the government as a customer, etc.
Like I said, I can just hit "Validate for section 508" in Visual Studio, there's your check list, that takes 10 minutes. Thats quite unlikely to keep the customer out of trouble, however...
If the requirement is to have an "accessible web site for our budget", then I 100% agree with you. Making a web site "mostly" accessible (assuming it doesn't have web APPLICATION requirements) is very very easy. Being compliant enough to get a government contract though? Better take out the books and get specialists involved.
Also, it was a bad analogy in the summary. Asking for good website design & and keeping ADA needs in consideration would be more like a diabetic demanding that sugar-free alternatives be made available.
Actually I think it was a decent analogy... Sugar-free alternatives are readily available. They are, however, a minority of the big corp's products, and most of the actual alternatives either come from TOTALLY DIFFERENT COMPANIES, or are TOTALLY DIFFERENT PRODUCTS.
So, as the CEO of a a big company... I could just let my competitors get the market if they want to, or just stick a phone number on my web site for blind people to call if they want to purchase something? That sounds good to me. (Actually, if I remember well, the later actually IS acceptable for ADA).
The web accessibility guidelines, if followed to the letter, will not give you an accessible web site. It will give you a "maybe". It is -very- easy to have a site that follows them all, is perfectly usuable by normal users, and are basically encrypted to the blind.
If you make a web site that actually IS accessible (not just following those silly guidelines), the web site may be decent, but you severly limit your options, and a LOT of features that are great to use in web applications cannot be used.
Also, it is interesting to note that the W3C guidelines are only one set of guidelines out of many, are incomplete for the various kinds of accessibility requirements that you may need (-very- incomplete), are not exactly matching section 508 (so those guidelines are not "enough", in many situations). Which was the point I was trying to make with my exagerated statement, which is: its easy as hell to make a web site accessible to a group of people (let say, the blinds). It is a whole other ball part to make a web site accessible.
Though even if you just target the blinds: without testing it it in screen readers, tough luck. Its like the W3C web standards in general. Sure, if I follow the standard to the letter, my site SHOULD work in all non-IE browsers. Thats cute in theory. In practice though? You'll need to test it around for the details, different defaults, bugs....
Well, everyone who doesn't think a website should look like a sophomore art magazine and work like an experimental X11 application from 1988
That pretty much includes 80% of every customer I ever had to deal with. Though a web app made with ExtJS looks a heck of a lot better than most -modern- desktop application:)
Section 508 isn't exactly something you can easily learn by heart however... it is rather...err.. "extensive". That doesn't help. I know Visual Studio has a validator for it...and it catches the obvious stuff... but some of it is rather up to interpretation...
It also means not using anything visually fancy or dynamic, and no script for anything important in the site (which isn't required for any document-style page, but for anything more application-like, well...)
All my web sites/web apps validate fine by the W3C standards/validator(s), and are event compliant by most official recommandations. Good luck reading some of them with a screen reader. Even a web site that is made with screen reader in mind... If you don't have a screenreader and regularly test it with, there's not a chance in hell (unless its a VERY simple web site) that it will be blind-friendly...
There's a lot more to do... anchors to allow screen readers to skip menus, css hidden tags to facilitate navigations, avoidance of html tables for complex page structure (I know you're not really supposed to...but good luck finding a reference to that on the W3C web site btw), semantic/ordered page structure...
Otherwise, the screen reader will be able to read it fine... It still won't make any sense.
No. The thing is, it never ends. And it ends up hurting "normal" people too. For the web, you need to make things accessible to screen readers, then you have to remember some blinds are poor and use shitty ones, so you have to deal with those cases too. Then people will want you to make the page browsable without a mouse because they may have physical limitations, then you have to be careful! People may be colorblind.
In the end, if I went on to point all of the things you have to do to comply with ADA and all similar recommendations, all you can really have is flat document-style black and white pages with rediculous font sizes made in raw XHTML/CSS and thats that.
We have technology, and the vast majority of people should be able to get the benefits of it. And making 6 versions of a web site is starting to be awkward, PLUS certain things on the web simply cannot be made that way either. And the customers want these things.
I'm all for doing what we can to help the disabled... but using the lowest common denominator will simply make everyone live their lives like if they were nearly disabled.
Thats if like, to fight poverty, we DID redistribute all of the wealth equally: there's not enough ressource for that, so instead of having some rich, some OK, and some poor, we'd all be poor, sick and dying. Thats not better. I know thats not the politically correct way of saying it, but seriously: lowering everyone to the level of the few unfortunate will get us nowhere.
I make all of my web sites with screen readers in mind as much as possible, using hidden tags to facilitate their navigation, making sure the basics are done (like alts on images), try to keep my markup semantic... but if a part of the web site would be more usuable to 99.999% of my customers using ajax/javascript tricks that will totally mess up a screen reader, well, I'm sorry. I'll make sure the first thing the screenreader will hit is a phone number they can call to get easier service. Beyond that... we're having issues with our deadlines as it is...
Exactly. And heck, even most of the good software that Linux DOES have was made for money (Firefox/Mozilla has a larger income stream than most companies I worked for...).
People have to sit down and let it sink in that money rules the world, even the Free Software world (with a couple of notable exceptions).
You're showing your age now:) One of my professors described that to me back then as his own experience... and last I was in school was almost a decade ago!:)
Definately, what I described is very close (but not quite) applied math. And its whats taught in the vast, vast majority of colleges (though its more like applied math + low level computer stuff). There are schools that teach Software Engineering, and thats much closer to what you're thinking of (I assume EE is electric engineering?). Software Engineering (in general), where its taught, is a wonderful program with a great balance between science and real world application, and that definately puts you in a situation where you can work after your first year or two.
I dont go by -my- definitions btw...just the definition of 95% of colleges.
Final note: I completly agree with you that its how it should be. However the reality of the field is otherwise, and it is why companies in business IT (business IT as opposed to computer science... so business XYZ vs Intel/Google/Microsoft/Nvidia and that kind of company) are reluctant to hire new graduates. As a general rule (let say 2/3rd of graduates, and some schools do not have that issue, though a lot of the "big names" do) need extensive trainings from their company afterward.
For example: Design Patterns and software architecture is virtually never taught, so my fiancee's company have dedicated times for design patterns classes during work hours. Microsoft hires people for "career" paths... they're good in CS out of college, but are trained for their real job.
The reason is that CS is, well, computer science... it is applied maths. Technically, if you like doing business IT, databases (except relational modeling), excel, etc, you're in the wrong branch. However, software development isn't a degree in most areas, so they go in CS instead (and companies have little choice but to hire CS students for business IT).
CS is simply not the field for that. A significant amount of students CAN do what you're thinking of, simply because, again technically, they're in the wrong degree to match their interests, but they have no choice. I feel that there should be a degree separate from CS for Software Development and Business processes for programmers (there are business process degrees, but not in a programming branch), and then those students could all do what you say after 3 months.
Thats, however, not the case. The amount of CS students who know the difference between a LEFT JOIN and an INNER JOIN when they graduate? Close to zero. It shouldn't be that way. The real world requires it to be different. But the programs are adapting very slowly. Unless a student enjoys learning things COMPLETLY unrelated to what they learn in class, they won't be able to take a job for a small business.
An internship for Intel though? Probably could after the first year.
no. the iphone sdk is used for both iphone and ipod touch, and apps must be signed for both (it's the same os, just slightly different hardware).
Thats good to know, thank you. However, for your comment in parenthesis.. that is true of Windows Mobile phones vs Pocket PCs for the most part, and they still behave differently when it comes to signing requirements (there's actually multiple levels of "security" that can be used for unsigned apps on Windows Mobile devices... though from what I see, the only ones that are really used is "free for all", and "locked down and requires a key", and the others mostly ignored).
Though with Windows Mobile, your application can run just peachy on a Pocket PC with the exact same code without being signed at all. Can iPhone apps run unsigned on an Ipod Touch? (actual, honest question).
Just another 2 cents... No, many people (most!) who go in CS don't have the skills to do work for a small company until very late in their degree...often never (and the company that hires train them. Very common among the likes of large comapnies). I personally worked on my own to get these skills faster (thus why I could get hired during my degree), but I actually started CS without even knowing how to boot a computer! (And 5 years after my degree, was a senior software architect, so I figure I did ok).
Considering how many colleges offer (as a mandatory class!) an introduction to computer class, Im guessing its common:) And again, with CS being about computer SCIENCE, in very very many colleges, you'll never get skills to work for a small company unless you learn stuff on your side (a -significant- amount of stuff).
Just for the sake of conversation... I'd be curious about what you consider a typical 2 year CS student able to do for a small company. Usually, companies the size of those you talk require networking, user support (With Windows...do they even show that in colleges anymore? Im jesting though, don't answer), excel stuff, small database driven applications (Having looked in the curriculum of a vast amount of US colleges, database/SQL and relational model is almost never taught), UI and front end development(again, rare in CS) and analysis (almost never taught aside in grad school).
So unless you have a decent amount of personal experience (even hacking up in Java/.NET in your spare time, or simply having industry knowledge), you're totally useless to them. So really, its not an option for the vast majority of CS majors:(
As a second sidenote: my girlfriend graduated a few years back, didn't do any internships (well, ok, she did Summer of Code), didn't do any side job (she didn't have ANY job experience... CS or otherwise), had a shitty GPA (2.5 or something), but she went to a "top school" (not a big name like MIT or Cal tech, but still a well known). She also skipped straight past all of the HR in all of her interviews (virtually only had interviews with engineers, architects and tech leads/PMs), and landed a 80k/year job a full year before she graduated (waiting for her for once she did).
Just to put things in perspective. And at her school, the above was the norm.
That does however mean you need to A) have the skills (CS degrees in general do not give you real world skills, or much of it, and when they do, its rarely in the first semesters... small 50 people companies don't need big O analysis of their algorythms), so you need to have the skills on your own. Many do. Many don't.
You also need to be doing well enough that you have time to juggle a job like that, AND that someone is willing to hire seasonal or part time. Harder than it sounds.
As opposed to going to a big name college, being picked up for a full time job several semesters before you even graduate for basically the company you want (give or take), and with a starting salary that will make your loans insignificant... The school does matter, for better or worse.
I'm not saying one method is better than the other. What i'm saying is that one's current situation should influence their choice heavily, because there are different advantages and drawbacks to going to a big name or not.
As I side note, I did do everything you suggested while I was in college... but it was seriously a -fraction- of the people there who were in a situation to do it. Millage may vary.
Honestly, the college does matter: for your -first- job. After that it doesn't. Employers who pick up straight out of college will actually care, especially clueless HR departments (but a lot of great companies have clueless HR).
As soon as you have a job or two behind you, then they won't even look at it (at least, most of em won't), and then you're right for the rest. But for the very first when you don't have experience aside internships? Sure does matter. At least, if you want to have all of the doors wide open.
They separated QA from programmer, but different surveys will draw the line of whats a QA and whats a programmer differently. Does the guy who writes unit tests count as a dev or not? Is the only one counting as a QA the test monkey who tries to run in every corners of the map? (Common).
And I specifically specified that the 7 digits were for -star- developers. Not everyday joe. Though even for not-quite-as-famous ones... The top tier of programmers in extremely successful companies will often run close to that... (well, "often", relatively speaking of course!). I know not-known-at-all Sharepoint developers running at 350$/hour on a payroll (not a contract! and in an area where cost of living is 1/3rd of what it is in NYC).
For the meaning of the average salary, you're quite correct. Unless of course, you live in a place where the cost of living is average:) (That said, I've worked for a significant amount of employers who based the salary of their employed base on the market's national average... so it is significant in that sense).
Im guessing for programmers they count everything, including QAs and low end staff.
The average salary in the US for a game dev/programmer (not counting interns, QAs, etc) is actually about 83.5k last I checked. (I forget the source).
That said, game development is a field of extremes. The star programmers will make 6 or (::gasps::) even 7 digit if they have parts in the company. Average joe "Programming games is cool!" out of school will have issues paying rent though. Developing business apps is a far more reliable source of income over a long period of time.
Seriously. This has always cracked me up... especially since nothing seems to be able to convince people that quality of wires (above a certain level... extremely shitty wires will have issues at extreme levels) is insignificant with digital and that it is so by design... I mean, its how the internet works, and god knows here's a lot of "noise" on the line, yet my text files sent over instant messengers are accurate:)
But nope, it doesn't get through. Monster HDMI cables >>>>> ALL, and nothing to do to convince anyone otherwise.
Go in the Media gallery, pick the category Web Docs/Videos, and watch "Unleashing the force".
A little after 1/3rd of the way through, they talk about DMM (Digital Molecular Matter, or something like that), and thats what im refering to. I'm sure its nowhere as sophisticated as they make it sound, but still, it looks cool:)
but Linux is highly specialized. There's no standardization for interaction
Thats why I feel that the future of Linux in user's hands is in the form of "appliance" type machines. Things like the EeePC, cellphones, Tivos-type things... we already have, and it works quite well. Now push it a notch further... a desktop machine with everything a user need, but locked down. Can't install or remove anything, except for the SD card or USB stick to store your data. Different models with different software for different people (and maybe like the EeePC, let people hack it up, but not by default).
Linux is -really- good at that kindda stuff. Linux desktops work great when they're preconfigured and you don't change em too much (which is when, for a regular user, all hell breaks loose).
I remember at my fiancee's college, most of the computer clusters were like that. Locked down desktop linux installs. It worked amazingly well. Since you couldn't screw it up, everything just worked, Mac-style. Very clean, all your files were saved on a network drive (as opposed to USB as I said above, but still), and you could install a limited amount of non-disruptive things.. if you messed up, you could just re-init it like you would a router.
There's nothing special about that...nothing that can't be done with Ubuntu and a few minutes/hours of tweaking. But if you sell that directly to users, you'll have a winner.
I know it was a joke, but encoding != document format:)
Anyway, if you ever had to deal with ISO standards before, you'd realise that what Microsoft did is the least of your worries. ISO, W3C, OASIS, ECMA... they all suck. They're all organisations that make "standards" by comittee, and while that sounds great in theory, in practice its more like:
Member A: "I want our next standard to have feature X" Member B: "No way, that would only further YOUR agenda and will destroy interop and/or makes its harder to implement for nothing! Instead, we should have feature Y, much better" Member A: "Nooo! That would only further YOUR agenda. Its even worse than X!" Member B: "Ok, what about this: you can have X, I can have Y, everyone's happy" Member C: "Wh...what? X and Y are mutually -exclusive-, you'll make it hell for -everyone- if we have both" Member A + B: "Two vs 1, we win, go to hell".
A lot of "standards", from all the stuff ISO has, to XHTML/XML/SOAP, stopping in between for things that are not so standards such as all of the accessibility acts and hell, the -law-, is made like this. And thats why it all sucks, and its all out of wack.
Compared to a lot of things that didn't cause so much of a stir, OOXML is a blessing... and thats not saying much. Point is, its nothing new, ISO, and most of the other standard bodies have always done this... this time it was just more visible because it was Microsoft... but anyone who tried to make a company ISO certified to various degree knows: you're better off going to IKEA for clear, sensible instructions.
Slashdot Mirror
I don't have doc to back it up, but as far as Ive read, with the write balancing features of SSDs, they can last a sick amount of time before the amount of rewrites finish them off. Also, they have one big big big advantage over HDDs: you know in advance that they are going to fail, quite reliably. So sure, it MAY (not all that true depending on usage anymore) fail faster than an HDD, but it won't fail by surprise on a saturday morning.
Make encryption illegal! /sarcasm
Its still fairly different however. Declaring whats IN the food is a big step down from changing the food itself. I guess the equivalent would be to label software with the available accessibility feature and accessibility level of the software itself.
Having all that info on the box of packaged software (or let say, on an about page in a web site), that is, much more info than we have now, letting disabled people make better choices about the products they use, and allowing search engines to better categorise web sites, would go a long way (let say, if it was enforced in a law that the declaration page/document has to conform with the web site itself, even if all it says is "We do not have any accessibility features"), I don't think anyone would complain. (And it would help the non-disabled).
I can make food that will be lethal to a group of people (peanuts!), I just have to say it is. I still can use peanuts/sugar/milk/whatever as much as I want, and don't have to make the slighest effort to make my food more "minority friendly". I just have to be honest about it.
Thats what I think the point in the article was.
Section 508 is fairly all or nothing though. If someone actually cared to be accessible for their customer base, they'd probably look more at actual accessibility recommendations, or the W3C accessibility specs. If you want Section 508, its probably to fullfil a legal requirements, you have the government as a customer, etc.
Like I said, I can just hit "Validate for section 508" in Visual Studio, there's your check list, that takes 10 minutes. Thats quite unlikely to keep the customer out of trouble, however...
If the requirement is to have an "accessible web site for our budget", then I 100% agree with you. Making a web site "mostly" accessible (assuming it doesn't have web APPLICATION requirements) is very very easy. Being compliant enough to get a government contract though? Better take out the books and get specialists involved.
Actually I think it was a decent analogy... Sugar-free alternatives are readily available. They are, however, a minority of the big corp's products, and most of the actual alternatives either come from TOTALLY DIFFERENT COMPANIES, or are TOTALLY DIFFERENT PRODUCTS.
So, as the CEO of a a big company... I could just let my competitors get the market if they want to, or just stick a phone number on my web site for blind people to call if they want to purchase something? That sounds good to me. (Actually, if I remember well, the later actually IS acceptable for ADA).
The web accessibility guidelines, if followed to the letter, will not give you an accessible web site. It will give you a "maybe". It is -very- easy to have a site that follows them all, is perfectly usuable by normal users, and are basically encrypted to the blind.
If you make a web site that actually IS accessible (not just following those silly guidelines), the web site may be decent, but you severly limit your options, and a LOT of features that are great to use in web applications cannot be used.
Also, it is interesting to note that the W3C guidelines are only one set of guidelines out of many, are incomplete for the various kinds of accessibility requirements that you may need (-very- incomplete), are not exactly matching section 508 (so those guidelines are not "enough", in many situations). Which was the point I was trying to make with my exagerated statement, which is: its easy as hell to make a web site accessible to a group of people (let say, the blinds). It is a whole other ball part to make a web site accessible.
Though even if you just target the blinds: without testing it it in screen readers, tough luck. Its like the W3C web standards in general. Sure, if I follow the standard to the letter, my site SHOULD work in all non-IE browsers. Thats cute in theory. In practice though? You'll need to test it around for the details, different defaults, bugs....
That pretty much includes 80% of every customer I ever had to deal with. Though a web app made with ExtJS looks a heck of a lot better than most -modern- desktop application
Section 508 isn't exactly something you can easily learn by heart however... it is rather...err.. "extensive". That doesn't help. I know Visual Studio has a validator for it...and it catches the obvious stuff... but some of it is rather up to interpretation...
It also means not using anything visually fancy or dynamic, and no script for anything important in the site (which isn't required for any document-style page, but for anything more application-like, well...)
All my web sites/web apps validate fine by the W3C standards/validator(s), and are event compliant by most official recommandations. Good luck reading some of them with a screen reader. Even a web site that is made with screen reader in mind... If you don't have a screenreader and regularly test it with, there's not a chance in hell (unless its a VERY simple web site) that it will be blind-friendly...
There's a lot more to do... anchors to allow screen readers to skip menus, css hidden tags to facilitate navigations, avoidance of html tables for complex page structure (I know you're not really supposed to...but good luck finding a reference to that on the W3C web site btw), semantic/ordered page structure...
Otherwise, the screen reader will be able to read it fine... It still won't make any sense.
No. The thing is, it never ends. And it ends up hurting "normal" people too. For the web, you need to make things accessible to screen readers, then you have to remember some blinds are poor and use shitty ones, so you have to deal with those cases too. Then people will want you to make the page browsable without a mouse because they may have physical limitations, then you have to be careful! People may be colorblind.
In the end, if I went on to point all of the things you have to do to comply with ADA and all similar recommendations, all you can really have is flat document-style black and white pages with rediculous font sizes made in raw XHTML/CSS and thats that.
We have technology, and the vast majority of people should be able to get the benefits of it. And making 6 versions of a web site is starting to be awkward, PLUS certain things on the web simply cannot be made that way either. And the customers want these things.
I'm all for doing what we can to help the disabled... but using the lowest common denominator will simply make everyone live their lives like if they were nearly disabled.
Thats if like, to fight poverty, we DID redistribute all of the wealth equally: there's not enough ressource for that, so instead of having some rich, some OK, and some poor, we'd all be poor, sick and dying. Thats not better. I know thats not the politically correct way of saying it, but seriously: lowering everyone to the level of the few unfortunate will get us nowhere.
I make all of my web sites with screen readers in mind as much as possible, using hidden tags to facilitate their navigation, making sure the basics are done (like alts on images), try to keep my markup semantic... but if a part of the web site would be more usuable to 99.999% of my customers using ajax/javascript tricks that will totally mess up a screen reader, well, I'm sorry. I'll make sure the first thing the screenreader will hit is a phone number they can call to get easier service. Beyond that... we're having issues with our deadlines as it is...
Exactly. And heck, even most of the good software that Linux DOES have was made for money (Firefox/Mozilla has a larger income stream than most companies I worked for...).
People have to sit down and let it sink in that money rules the world, even the Free Software world (with a couple of notable exceptions).
You're showing your age now :) One of my professors described that to me back then as his own experience... and last I was in school was almost a decade ago! :)
Definately, what I described is very close (but not quite) applied math. And its whats taught in the vast, vast majority of colleges (though its more like applied math + low level computer stuff). There are schools that teach Software Engineering, and thats much closer to what you're thinking of (I assume EE is electric engineering?). Software Engineering (in general), where its taught, is a wonderful program with a great balance between science and real world application, and that definately puts you in a situation where you can work after your first year or two.
I dont go by -my- definitions btw...just the definition of 95% of colleges.
Final note: I completly agree with you that its how it should be. However the reality of the field is otherwise, and it is why companies in business IT (business IT as opposed to computer science... so business XYZ vs Intel/Google/Microsoft/Nvidia and that kind of company) are reluctant to hire new graduates. As a general rule (let say 2/3rd of graduates, and some schools do not have that issue, though a lot of the "big names" do) need extensive trainings from their company afterward.
For example: Design Patterns and software architecture is virtually never taught, so my fiancee's company have dedicated times for design patterns classes during work hours. Microsoft hires people for "career" paths... they're good in CS out of college, but are trained for their real job.
The reason is that CS is, well, computer science... it is applied maths. Technically, if you like doing business IT, databases (except relational modeling), excel, etc, you're in the wrong branch. However, software development isn't a degree in most areas, so they go in CS instead (and companies have little choice but to hire CS students for business IT).
CS is simply not the field for that. A significant amount of students CAN do what you're thinking of, simply because, again technically, they're in the wrong degree to match their interests, but they have no choice. I feel that there should be a degree separate from CS for Software Development and Business processes for programmers (there are business process degrees, but not in a programming branch), and then those students could all do what you say after 3 months.
Thats, however, not the case. The amount of CS students who know the difference between a LEFT JOIN and an INNER JOIN when they graduate? Close to zero. It shouldn't be that way. The real world requires it to be different. But the programs are adapting very slowly. Unless a student enjoys learning things COMPLETLY unrelated to what they learn in class, they won't be able to take a job for a small business.
An internship for Intel though? Probably could after the first year.
Thats good to know, thank you. However, for your comment in parenthesis.. that is true of Windows Mobile phones vs Pocket PCs for the most part, and they still behave differently when it comes to signing requirements (there's actually multiple levels of "security" that can be used for unsigned apps on Windows Mobile devices... though from what I see, the only ones that are really used is "free for all", and "locked down and requires a key", and the others mostly ignored).
Though with Windows Mobile, your application can run just peachy on a Pocket PC with the exact same code without being signed at all. Can iPhone apps run unsigned on an Ipod Touch? (actual, honest question).
Just another 2 cents... No, many people (most!) who go in CS don't have the skills to do work for a small company until very late in their degree...often never (and the company that hires train them. Very common among the likes of large comapnies). I personally worked on my own to get these skills faster (thus why I could get hired during my degree), but I actually started CS without even knowing how to boot a computer! (And 5 years after my degree, was a senior software architect, so I figure I did ok).
:) And again, with CS being about computer SCIENCE, in very very many colleges, you'll never get skills to work for a small company unless you learn stuff on your side (a -significant- amount of stuff).
:(
Considering how many colleges offer (as a mandatory class!) an introduction to computer class, Im guessing its common
Just for the sake of conversation... I'd be curious about what you consider a typical 2 year CS student able to do for a small company. Usually, companies the size of those you talk require networking, user support (With Windows...do they even show that in colleges anymore? Im jesting though, don't answer), excel stuff, small database driven applications (Having looked in the curriculum of a vast amount of US colleges, database/SQL and relational model is almost never taught), UI and front end development(again, rare in CS) and analysis (almost never taught aside in grad school).
So unless you have a decent amount of personal experience (even hacking up in Java/.NET in your spare time, or simply having industry knowledge), you're totally useless to them. So really, its not an option for the vast majority of CS majors
As a second sidenote: my girlfriend graduated a few years back, didn't do any internships (well, ok, she did Summer of Code), didn't do any side job (she didn't have ANY job experience... CS or otherwise), had a shitty GPA (2.5 or something), but she went to a "top school" (not a big name like MIT or Cal tech, but still a well known). She also skipped straight past all of the HR in all of her interviews (virtually only had interviews with engineers, architects and tech leads/PMs), and landed a 80k/year job a full year before she graduated (waiting for her for once she did).
Just to put things in perspective. And at her school, the above was the norm.
That does however mean you need to A) have the skills (CS degrees in general do not give you real world skills, or much of it, and when they do, its rarely in the first semesters... small 50 people companies don't need big O analysis of their algorythms), so you need to have the skills on your own. Many do. Many don't.
You also need to be doing well enough that you have time to juggle a job like that, AND that someone is willing to hire seasonal or part time. Harder than it sounds.
As opposed to going to a big name college, being picked up for a full time job several semesters before you even graduate for basically the company you want (give or take), and with a starting salary that will make your loans insignificant... The school does matter, for better or worse.
I'm not saying one method is better than the other. What i'm saying is that one's current situation should influence their choice heavily, because there are different advantages and drawbacks to going to a big name or not.
As I side note, I did do everything you suggested while I was in college... but it was seriously a -fraction- of the people there who were in a situation to do it. Millage may vary.
Honestly, the college does matter: for your -first- job. After that it doesn't. Employers who pick up straight out of college will actually care, especially clueless HR departments (but a lot of great companies have clueless HR). As soon as you have a job or two behind you, then they won't even look at it (at least, most of em won't), and then you're right for the rest. But for the very first when you don't have experience aside internships? Sure does matter. At least, if you want to have all of the doors wide open.
They separated QA from programmer, but different surveys will draw the line of whats a QA and whats a programmer differently. Does the guy who writes unit tests count as a dev or not? Is the only one counting as a QA the test monkey who tries to run in every corners of the map? (Common).
:) (That said, I've worked for a significant amount of employers who based the salary of their employed base on the market's national average... so it is significant in that sense).
And I specifically specified that the 7 digits were for -star- developers. Not everyday joe. Though even for not-quite-as-famous ones... The top tier of programmers in extremely successful companies will often run close to that... (well, "often", relatively speaking of course!). I know not-known-at-all Sharepoint developers running at 350$/hour on a payroll (not a contract! and in an area where cost of living is 1/3rd of what it is in NYC).
For the meaning of the average salary, you're quite correct. Unless of course, you live in a place where the cost of living is average
Im guessing for programmers they count everything, including QAs and low end staff.
The average salary in the US for a game dev/programmer (not counting interns, QAs, etc) is actually about 83.5k last I checked. (I forget the source).
That said, game development is a field of extremes. The star programmers will make 6 or (::gasps::) even 7 digit if they have parts in the company. Average joe "Programming games is cool!" out of school will have issues paying rent though. Developing business apps is a far more reliable source of income over a long period of time.
Seriously. This has always cracked me up... especially since nothing seems to be able to convince people that quality of wires (above a certain level... extremely shitty wires will have issues at extreme levels) is insignificant with digital and that it is so by design... I mean, its how the internet works, and god knows here's a lot of "noise" on the line, yet my text files sent over instant messengers are accurate :)
But nope, it doesn't get through. Monster HDMI cables >>>>> ALL, and nothing to do to convince anyone otherwise.
I hate Flash sites, but:
:)
http://www.lucasarts.com/games/theforceunleashed/
Go in the Media gallery, pick the category Web Docs/Videos, and watch "Unleashing the force".
A little after 1/3rd of the way through, they talk about DMM (Digital Molecular Matter, or something like that), and thats what im refering to. I'm sure its nowhere as sophisticated as they make it sound, but still, it looks cool
Thats why I feel that the future of Linux in user's hands is in the form of "appliance" type machines. Things like the EeePC, cellphones, Tivos-type things... we already have, and it works quite well. Now push it a notch further... a desktop machine with everything a user need, but locked down. Can't install or remove anything, except for the SD card or USB stick to store your data. Different models with different software for different people (and maybe like the EeePC, let people hack it up, but not by default).
Linux is -really- good at that kindda stuff. Linux desktops work great when they're preconfigured and you don't change em too much (which is when, for a regular user, all hell breaks loose).
I remember at my fiancee's college, most of the computer clusters were like that. Locked down desktop linux installs. It worked amazingly well. Since you couldn't screw it up, everything just worked, Mac-style. Very clean, all your files were saved on a network drive (as opposed to USB as I said above, but still), and you could install a limited amount of non-disruptive things.. if you messed up, you could just re-init it like you would a router.
There's nothing special about that...nothing that can't be done with Ubuntu and a few minutes/hours of tweaking. But if you sell that directly to users, you'll have a winner.
I know it was a joke, but encoding != document format :)
Anyway, if you ever had to deal with ISO standards before, you'd realise that what Microsoft did is the least of your worries. ISO, W3C, OASIS, ECMA... they all suck. They're all organisations that make "standards" by comittee, and while that sounds great in theory, in practice its more like:
Member A: "I want our next standard to have feature X"
Member B: "No way, that would only further YOUR agenda and will destroy interop and/or makes its harder to implement for nothing! Instead, we should have feature Y, much better"
Member A: "Nooo! That would only further YOUR agenda. Its even worse than X!"
Member B: "Ok, what about this: you can have X, I can have Y, everyone's happy"
Member C: "Wh...what? X and Y are mutually -exclusive-, you'll make it hell for -everyone- if we have both"
Member A + B: "Two vs 1, we win, go to hell".
A lot of "standards", from all the stuff ISO has, to XHTML/XML/SOAP, stopping in between for things that are not so standards such as all of the accessibility acts and hell, the -law-, is made like this. And thats why it all sucks, and its all out of wack.
Compared to a lot of things that didn't cause so much of a stir, OOXML is a blessing... and thats not saying much. Point is, its nothing new, ISO, and most of the other standard bodies have always done this... this time it was just more visible because it was Microsoft... but anyone who tried to make a company ISO certified to various degree knows: you're better off going to IKEA for clear, sensible instructions.