How about taking that idea one further and modeling the voting machines after well tested technology, both for usability and security...ATMs. They have an interface that works, plus most people are already familiar with it. They have a paper trail in addition to digitally stored state (redundancy should be a no-brainer for this machine). Plus they are secured against tampering.
You may or may not find the scriptable language that you're looking for. Your real problem is minimizing the effort required to produce custom programs, right? How about define a few narrowly scoped types of courseware that the teachers want and create some wizards that guide them through building it?
Instead of making them script a multiple choice test, have an option in your TestBuilder application to make one. Then provide them with a wizard that steps them through each step, allows previews, scoring criteria, etc. Put your effort into making it really easy to use by _working with the end-users_ and giving them what they want. I bet none of them will ask for a scripting language.
Find one or two teachers that are particularly interested in this software and get them to sign on as the "domain experts". Set their expectations properly, that is let them know the first two or three iterations of the software will have some usability issues and that it's up to them to work with the developers to iron things out for everybody.
Bah, I prefer a time and materials contract at a decent hourly rate. That way the management side of things (handling change orders, scope changes, compensating for the lack of capability of others, etc) is taken care of. The client carries the risk of not having their s@!# together, not me, and if they don't I don't have to suddenly go into "reevaluation mode", I can just pull out the solution to the new problem and charge for my time.
Giving accurate estimates for how long work will take and then meeting those estimates is still an important part of the job, but this arrangement really cuts down on the non-paying and less interesting parts of my work. The downside is I don't get to rip anyone off by overcharging for my work. But I'm OK with that, I'm just here to solve peoples business problems and make them more productive.
Fixed bid contracts *do* work well when the work is very specifically defined, but I find that the customer that understands what they want that well is pretty rare. Kudos to you if you can find them.
Both extremes are wrong. Maintaining old static applications indefinitely is wrong. Periodic discard and replacement is wrong. My final conclusion was that old applications need to be rewritten and re-implemented and expanded and modernized gradually. If we re-write or re-implement 10% of the code every year, then none of the parts get to be more than 10 years old. We also deliberately blur the boundaries between old and new applications and the boundaries between developers and maintainers.
When to rewrite isn't a function of the codes age, it's a function of cost. How much will it cost to work around code versus rewrite it and when will the return on investment be made. Since most engineers work for businesses the proper evaluation of this will involve business issues outside of the scope of the technical work. Managing a business isn't really that much different than engineering, at least from this self-employed engineers point of view.
One other thing, when our gradual rewrites of old code reach the point where everything is fully expressed as objects, then the burden of rewrites and maintenance should be drastically reduced forever after. Isn't that the promise of objects? Expandability? Adaptability? Any large application well founded on objects should be able to morph itself into any future application one little bit at a time.
That's the (unkept) promise of every new programming paradigm. OO *is* great. The point is to make abstraction and isolation easier than it was in procedural programming. It's a godsend in large applications for this reason, but it still has problems. One is that abstractions don't start perfect and as other things in the system change the likelihood of having to do extra ugly work-arounds or a rewrite (see previous paragraph) is high. What OO does do for you is that when it it is used well it helps the developer have to think about fewer issues while programming, which should increase productivity.
I've worked as a contractor in this situation. My client had a machine dedicated as the build machine which I would connect to with vnc over ssh.
One HUGE advantage of this is that the client can easily build the project himself. If I decided to use a new library I have to set it up on the build machine for them. If I got hit by a bus tomorrow while carrying my workstation the client would have everything that they need to continue on with the project (minus one software developer).
How about using a website capture tool to grab the dynamic development site and then publish the static site to the host? There are lots of website capture tools, some OSS.
That's not quite one click, but some of them may be scriptable which could make it really close to one click.
My advice is to get a degree to help sell yourself. Yes, it might not be all fun and it might not be just, but if you want to have a fun job as a programmer it will help you achieve your goal. Nothing talks like results.
My other advice for you specifically, due to your autism, is to seek jobs with large engineering departments. At a small company you have to be a jack-of-all-trades. At a larger company you are more likely to be able to fit into a niche while other people handle the sides of an engineering job that you may find more difficult. For example, working with business people to create a workflow model documentation that engineering will later have to implement.
At a larger place you may be able to find a position where you get handed a well defined specification and told to go at it. Most likely this will be at engineering firms that work with hardware. The slower hardware development life-cycle forces the code specifications to be more rigidly defined.
Teach them how to hold usability tests in a cost effective fashion, covering both the math/statistics side as well as methodologies. This is essential to the success of any complex website.
Have them learn to look at the features of the available technologies and weigh how their strengths and weaknesses apply to the end user interface (this is mostly going to be looking at cost, development time, maintainability, and reliability).
This class should really focus on the life cycle of evaluating a websites effectiveness, how to improve it, how to design new features to solve business problems, test those features on users, etc.
These lessons will be timeless and get people ready to do real work in the real world. The technology behind it all will constantly change. The students can be expected to pick a good deal of it up on the side, just as they will be expected to during their professional lives. That's the kind of practice they need.
If you want to get the best prep for transfering into an undergraduate CS program you can take every math course possible during your first 2 years. Other math or logic heavy classes such as physics and philosophy will help tune your brain and you have a much better chance of their credit transfering, putting you ahead financially and time-wise.
Sure, take a couple classes that include some programming to see if you like it, but don't expect that class to count towards your eventual BS degree.
I've found myself in a similar situation. I had tons of pet projects when I was in school, even when I co-op'd, but as my development jobs got more intense and difficult I didn't have the same urge to go do more of the same at home. What I found works for me and still scratches the somewhat the same itch is to read GOOD computer books, the classics in the field. This excersizes your brain in a slightly different way, and will give you food for thought as you reflect on all the development you've been doing during the day.
I'll leave you with this thought. While some of the greatest hackers spend nearly all of their time hacking, this might not be the path that leads you to be the best hacker you can be. Myself (not the greatest hacker in the world, but I'm no slouch), I find that I program better when I've come back refreshed from other activities, such as playing musical instruments, excersizing, reading a book about a whole new field, etc.
Just some thoughts, don't feel guilty about not programming all the time, give yourself some space from it so you can enjoy it!
When it comes down to it, all a computer can do is manipulate numbers.
Here you haven't thought it through far enough. All a computer can do is manipulate phenomenon that we identify as electrons and through that manipulation cause a variety of other physically observable events. It is your mind that makes the numbers.
Computer science is about math yes, but it's not about computers, really. Those particular machines are studied more intently in computer engineering and software engineering.
It should be noted that most CS degrees do teach a lot of software engineering classes, there is a large intersection of material amoung the various computer oriented degrees.
You know, I think the poor guy has paid his debt to society. Yes, he made some big mistakes, but now he's moving on and he IS earning an honest living.
What would you do in his present situation? Go work a crappy job or make the most of the situation? I know what I'd do and I don't blame him for doing the same.
The guy is not the hero people make him out to be, but he also does not need negative creeps kicking him down any more. His debt has been paid, grow up.
If you tell everyone to avoid street 1, out of 2 streets, then you flood street 2. If it can be clustered or load balanced (beowulf?) so that people are distributed evenly on alternate routes.. this could be REALLY viable.
What would be really viable is if *only I* knew about the traffic free route. Now THAT would be technology.
I wasn't learning anything, and I had too much time available to me to play games
Um, who's fault is that? There's not much else one requires to learn other than time, you've already said you could get free trial equipment to play with and I have to imagine that there was plenty of handy reading material at a university.
Somehow, people survived for thousands of years without air conditioning. Could it be that "modern" housing design, where one style suits the entire country, isn't the best idea?
No, it's just that we turned into a bunch of pampered wooses. Seriously.
Imagine if one day, in the distant future, EVERYONE was required to have one of these...even if you have no intention of doing it, could lead to investigations
I tell you what, the size of the government required to investigate every potentially subversive thought *I alone* have would be so large as to collapse under it's own weight.
I couldn't understand why used Martin's sold for $3000-$3500 in music shops until I picked one up and played it -- it was worth the price!
And that's why I make sure I don't play one of those when I'm checking out local used gear. That's an itch I can't afford to have! Best not to know what I'm missing.
Slashdot Mirror
How about taking that idea one further and modeling the voting machines after well tested technology, both for usability and security...ATMs. They have an interface that works, plus most people are already familiar with it. They have a paper trail in addition to digitally stored state (redundancy should be a no-brainer for this machine). Plus they are secured against tampering.
You may or may not find the scriptable language that you're looking for. Your real problem is minimizing the effort required to produce custom programs, right? How about define a few narrowly scoped types of courseware that the teachers want and create some wizards that guide them through building it?
Instead of making them script a multiple choice test, have an option in your TestBuilder application to make one. Then provide them with a wizard that steps them through each step, allows previews, scoring criteria, etc. Put your effort into making it really easy to use by _working with the end-users_ and giving them what they want. I bet none of them will ask for a scripting language.
Find one or two teachers that are particularly interested in this software and get them to sign on as the "domain experts". Set their expectations properly, that is let them know the first two or three iterations of the software will have some usability issues and that it's up to them to work with the developers to iron things out for everybody.
Bah, I prefer a time and materials contract at a decent hourly rate. That way the management side of things (handling change orders, scope changes, compensating for the lack of capability of others, etc) is taken care of. The client carries the risk of not having their s@!# together, not me, and if they don't I don't have to suddenly go into "reevaluation mode", I can just pull out the solution to the new problem and charge for my time.
Giving accurate estimates for how long work will take and then meeting those estimates is still an important part of the job, but this arrangement really cuts down on the non-paying and less interesting parts of my work. The downside is I don't get to rip anyone off by overcharging for my work. But I'm OK with that, I'm just here to solve peoples business problems and make them more productive.
Fixed bid contracts *do* work well when the work is very specifically defined, but I find that the customer that understands what they want that well is pretty rare. Kudos to you if you can find them.
Both extremes are wrong. Maintaining old static applications indefinitely is wrong. Periodic discard and replacement is wrong. My final conclusion was that old applications need to be rewritten and re-implemented and expanded and modernized gradually. If we re-write or re-implement 10% of the code every year, then none of the parts get to be more than 10 years old. We also deliberately blur the boundaries between old and new applications and the boundaries between developers and maintainers.
When to rewrite isn't a function of the codes age, it's a function of cost. How much will it cost to work around code versus rewrite it and when will the return on investment be made. Since most engineers work for businesses the proper evaluation of this will involve business issues outside of the scope of the technical work. Managing a business isn't really that much different than engineering, at least from this self-employed engineers point of view.
One other thing, when our gradual rewrites of old code reach the point where everything is fully expressed as objects, then the burden of rewrites and maintenance should be drastically reduced forever after. Isn't that the promise of objects? Expandability? Adaptability? Any large application well founded on objects should be able to morph itself into any future application one little bit at a time.
That's the (unkept) promise of every new programming paradigm. OO *is* great. The point is to make abstraction and isolation easier than it was in procedural programming. It's a godsend in large applications for this reason, but it still has problems. One is that abstractions don't start perfect and as other things in the system change the likelihood of having to do extra ugly work-arounds or a rewrite (see previous paragraph) is high. What OO does do for you is that when it it is used well it helps the developer have to think about fewer issues while programming, which should increase productivity.
"L-Systems" mean the same kind of L_System used for autogenerating graphics?
I've seen some neat organic looking images come from that.
I've worked as a contractor in this situation. My client had a machine dedicated as the build machine which I would connect to with vnc over ssh.
One HUGE advantage of this is that the client can easily build the project himself. If I decided to use a new library I have to set it up on the build machine for them. If I got hit by a bus tomorrow while carrying my workstation the client would have everything that they need to continue on with the project (minus one software developer).
Jon
How about using a website capture tool to grab the dynamic development site and then publish the static site to the host? There are lots of website capture tools, some OSS.
That's not quite one click, but some of them may be scriptable which could make it really close to one click.
My advice is to get a degree to help sell yourself. Yes, it might not be all fun and it might not be just, but if you want to have a fun job as a programmer it will help you achieve your goal. Nothing talks like results.
My other advice for you specifically, due to your autism, is to seek jobs with large engineering departments. At a small company you have to be a jack-of-all-trades. At a larger company you are more likely to be able to fit into a niche while other people handle the sides of an engineering job that you may find more difficult. For example, working with business people to create a workflow model documentation that engineering will later have to implement.
At a larger place you may be able to find a position where you get handed a well defined specification and told to go at it. Most likely this will be at engineering firms that work with hardware. The slower hardware development life-cycle forces the code specifications to be more rigidly defined.
Good luck.
Teach them how to hold usability tests in a cost effective fashion, covering both the math/statistics side as well as methodologies. This is essential to the success of any complex website.
Have them learn to look at the features of the available technologies and weigh how their strengths and weaknesses apply to the end user interface (this is mostly going to be looking at cost, development time, maintainability, and reliability).
This class should really focus on the life cycle of evaluating a websites effectiveness, how to improve it, how to design new features to solve business problems, test those features on users, etc.
These lessons will be timeless and get people ready to do real work in the real world. The technology behind it all will constantly change. The students can be expected to pick a good deal of it up on the side, just as they will be expected to during their professional lives. That's the kind of practice they need.
Have you seen _The Karate Kid_?
If you want to get the best prep for transfering into an undergraduate CS program you can take every math course possible during your first 2 years. Other math or logic heavy classes such as physics and philosophy will help tune your brain and you have a much better chance of their credit transfering, putting you ahead financially and time-wise.
Sure, take a couple classes that include some programming to see if you like it, but don't expect that class to count towards your eventual BS degree.
I've found myself in a similar situation. I had tons of pet projects when I was in school, even when I co-op'd, but as my development jobs got more intense and difficult I didn't have the same urge to go do more of the same at home. What I found works for me and still scratches the somewhat the same itch is to read GOOD computer books, the classics in the field. This excersizes your brain in a slightly different way, and will give you food for thought as you reflect on all the development you've been doing during the day.
I'll leave you with this thought. While some of the greatest hackers spend nearly all of their time hacking, this might not be the path that leads you to be the best hacker you can be. Myself (not the greatest hacker in the world, but I'm no slouch), I find that I program better when I've come back refreshed from other activities, such as playing musical instruments, excersizing, reading a book about a whole new field, etc.
Just some thoughts, don't feel guilty about not programming all the time, give yourself some space from it so you can enjoy it!
I must agree that "Charbux" coffee is extremely over-roasted.
It's a french roast, it's supposed to be like that. Compare starbucks french roast to any other french roast, it's not that different.
You are most definitely right, CS is all math.
Pretty much.
When it comes down to it, all a computer can do is manipulate numbers.
Here you haven't thought it through far enough. All a computer can do is manipulate phenomenon that we identify as electrons and through that manipulation cause a variety of other physically observable events. It is your mind that makes the numbers.
Computer science is about math yes, but it's not about computers, really. Those particular machines are studied more intently in computer engineering and software engineering.
It should be noted that most CS degrees do teach a lot of software engineering classes, there is a large intersection of material amoung the various computer oriented degrees.
You know, I think the poor guy has paid his debt to society. Yes, he made some big mistakes, but now he's moving on and he IS earning an honest living.
What would you do in his present situation? Go work a crappy job or make the most of the situation? I know what I'd do and I don't blame him for doing the same.
The guy is not the hero people make him out to be, but he also does not need negative creeps kicking him down any more. His debt has been paid, grow up.
I prefer a less formal design approach best exemplified by this guiding principle:
;)
A design should be written on the smallest napkin possible, but not smaller.
Anything more than that is gravy.
If you tell everyone to avoid street 1, out of 2 streets, then you flood street 2. If it can be clustered or load balanced (beowulf?) so that people are distributed evenly on alternate routes.. this could be REALLY viable.
What would be really viable is if *only I* knew about the traffic free route. Now THAT would be technology.
I wasn't learning anything, and I had too much time available to me to play games
Um, who's fault is that? There's not much else one requires to learn other than time, you've already said you could get free trial equipment to play with and I have to imagine that there was plenty of handy reading material at a university.
Somehow, people survived for thousands of years without air conditioning. Could it be that "modern" housing design, where one style suits the entire country, isn't the best idea?
No, it's just that we turned into a bunch of pampered wooses. Seriously.
Imagine if one day, in the distant future, EVERYONE was required to have one of these...even if you have no intention of doing it, could lead to investigations
I tell you what, the size of the government required to investigate every potentially subversive thought *I alone* have would be so large as to collapse under it's own weight.
Let's bring 'em down!
I cant believe its gonna miss! Now i cant throw my wicked end-of-the-world orgy-party! *sigh*
I don't know, that never stopped the SubGenius from having their annual-end-of-the-world orgies.
Give me a dart and a lawyer and I'll prove you wrong.
Also note that the size of the tumbleweed rover proposed for use on mars has a 6 meter diameter, basically it's a bigger sail than the earth model.
I couldn't understand why used Martin's sold for $3000-$3500 in music shops until I picked one up and played it -- it was worth the price!
And that's why I make sure I don't play one of those when I'm checking out local used gear. That's an itch I can't afford to have! Best not to know what I'm missing.
You keep using this word anymore. I do not think it means what you think it means.