- AMD was on top of the world with Opteron / AMD64 - Intel was losing everywhere it went. You'd be hard-pressed to find an Internet / financial shop *not* buying AMD - But Intel responded with Merom / Core2Duo. That mostly closed the gap, though initially the memory subsystem was still inferior - Had AMD met expectations with the follow-on part (Bulldozer), there is no reason they could not have continued to win - But in my mind, their ATi acquisition initiated their downfall. They became schizophrenic.
To beat Intel (like most market leaders) you have to have a non-trivial advantage. When AMD had one, they kicked Intel's ass to the point that they severely altered Intel's roadmap. When they no longer had one, they lost.
Last time I looked at buying SF property potentially to rent out I was scared off by the regulations the city imposes on landlords. How can anyone sleep at night knowing your tenants are more in control than you are? I'm going to bet I'm not the only one with this impression, which would mean given a choice individual owners will sell long before they'll risk getting into the 'affordable rent' game. Same dynamic may be why the developments are condos, not rentals.
So why no mention of the city's propensity to make being a landlord miserable?
Your post summarizes beautifully the fatal flaw with MS tools and Market-tectures.
They work great if you go down MS's beaten path.
They leave you in a lurch if you need to go off the path.
All along, Microsoft will trumpet the technology as the answer to everything and cause countless developers pain as they follow the same unreliable path only to fnid they can't ship anything interesting. It was a big eye-opener to me the day I realized that Microsoft told everyone to use MFC but shipped no MFC-based apps themselves.
Got to admit, though, that C# is pretty nice as a language.
I'm sure there are some brilliant nurses. I'm sure nurses are responsible for countless saved lives. I'm sure doctors could learn a lot if they spent time doing nursing. I'm sure that some nurses know more about practical human health mechanisms than some doctors do. But I have never heard of a doctor that got his M.D., then did nursing, then became a doctor.
I ride a bike 4-5x a week to work. My car often goes weeks - maybe a month - between drives. And I encourage everyone to do the same if they can.
But not because of CO2. CO2 is a trace gas comprising only.04% of the atmosphere. Humans emit only a small portion of the world's CO2, but it seems we must believe that the Earth can absorb and produce finite, unchanging amounts of CO2 such that any perturbation is disastrous. The climate data is not consistent, nor is the science predictive. Yes, pollution - as in things that dirty other things - is abhorrent. Yes, saving energy and money is great and we all need more exercise. But the CO2 obsession is a cult that insists everyone join.
...that the creator of C++ runs Windows. C++ is a barely-stomach-able OO implementation, and Windows is a barely-usable GUI & OS. I look forward to using neither, but often must do so.
Microsoft's problem is with software more than hardware. The Windows phones may not have been as slick as an iPhone, but the hardware sufficed in terms of durability, phone quality and screen. What did not suffice was/is the software, and I believe that is why Apple walked away with the smart phone market despite Microsoft's much earlier presence. Microsoft has existed in an environment for 30+ years in which poor or barely-passable software could lead to tremendous success. Those days are THANKFULLY over. Do they realize this? Or, just as Ringo Star claimed to be the world's best drummer (by association to the worlds most prevalent band) do Microsoft managers walk around confidently trumpeting that Windows is the greatest OS in the world?
The day that I can buy a non-Mac laptop that has a slick, polished, Unix-ish OS in a quiet, pleasing metal form factor is the day I even think about caring about cheaper.
My job is to code. I had a neighbor that worked in a paint factory. He got hurt, and went to school to learn software. Pretty soon, he's coming home talking to me about UML / Booch and other esoteric OOP topics that I barely had a grasp of. I thought, gee, wow, it is amazing to see how fast a blue collar guy can come up to speed. But then it all kind of fizzled out. And I wondered for a long time why that was. How could someone get so far and be able to recite complex comp-sci topics and then give up?
And the answer is, there are (at least) 2 components to being a s/w developer. You need to be able to think and analyze and understand. But you also need to be able to - and more importantly - slog it out to make it work. And it's that slogging out that - at least in my experience - is where you truly grok the information and become effective and confident and useful.
So, that's why I say 'beware'. You might be having one of those 'I know Kung Fu' moments, because you can understand your textbook. And that is a wonderful thing. But unless you cement that knowledge by actually using it - and in particular, by learning to debug hairy problems - it is at risk of floating away.
I mean, I could pick up a cutting edge cardiology book and, given a few weeks, recite and even talk intelligently about topics that most doctors wouldn't know. I could probably impress virtually all of the population with this knowledge. But, I would be a long way from being someone anyone would hire to cut someone open. Balance your book knowledge with practical knowledge. You need a residency.
Maybe it will be useless. Maybe you'll switch careers. Maybe you'd make more money working now than the increased pay later. Heck, in my case, I don't even have a bachelor's in CS and I do pretty well.
On the other hand, after a certain age it gets a lot harder to go back. Your time is suddenly spoken for by family, career, etc. Why take 3-4 years to finish part time at 2012 tuitions when you can knock it out in 1-2 years now?
In your posting you talked about adding ducts to steal A/C from a second unit. To work decently you would need to not only add a new output duct 'run', but also a new return 'run' (that is, unless the 2 units share a network of return ducting).
I remember how exciting the touch screens seemed in the late 80s, but when using them reality set in - you quickly fatigue holding your arm up to touch a screen.
Plus, with any user interface people need a certain confidence in correspondence between what they do and what happens. When you push a button, you KNOW it got pressed. If you push a joystick left, you KNOW you're going left. That 'payoff' is like a contract between you and the machine that goes favorably. But if pressing the screen where you believe you need to press may or may not do what you want, that contract gets shaky. Especially since there's no click or motion to reinforce what you're doing. This, by the way, is why I think 'free space' VR controllers never caught on...at least until the WII.
Still, software can create cues to take the place of physicality and have 'grease' to avoid common miscues. Plus, having the screen be horizontal reduces the fatigue.
But in the end, as archaic as the keyboard seems compared to touch and speech, it really is an incredibly expressive and low-energy-requirement device.
Now that I've been out of school for (gasp) 17 years I see it this way.
Before I talk engineering, I want to say this: the most valuable class for me in all my college career was the one that taught me how to write. A so-so engineer with excellent communication skills can go a lot further than the reverse.
Anyway...at one time because of demand for software people you really could squeak by without a very good degree or any degree at all. That's probably still true to some lesser extent. And I have to say that new grads fail to impress me with basic software *engineering* (debugging, organization, maintainability) skills.
But I *still* highly recommend you get the best degree you can get.
For engineering, my take is that there are like 4 different tiers of schools.
Tier 1) A degree from MIT and to a lesser extent Stanford is a gift that keeps on giving. For the rest of your career you will be referred to as 'that guy who went to MIT'. Even if you barely passed. Just like Gordon Freeman. Out on the west coast having a Stanford degree is about on par. But MIT seems to have students actually build stuff, which is not only fun but which also translates into people with better basic engineering skills. So, if you can manage to get in and graduate you will never regret it.
Tier 2) Then you have the rest of the 'top 10' or 'top 15' ranked schools. I'm out of date, but I'm thinking like Berkeley, RPI, WPI, Univ of Illinois Urbana, Michigan, some (but certainly not all!) of the ivy league schools and others my ignorance causes me to miss. Schools in this league are not only really helpful in getting your first job but more importantly they will WORK YOU. I went to what I call a 'tier 3)' school first for 2 years before going to a 'tier 2)' school, and boy was it a different level of intensity. That gives you skills and confidence.
The funny thing is that sometimes these schools rank above or darn close to the 'Tier 1)' schools. I went to Urbana, and I know there were years where it was #2 or #1. And from a curriculum standpoint you probably are getting just as good of an education. But - regardless of what your guidance counselor or parents tell you - even if MIT was consistently ranked #9, having that MIT degree buys you something that the Urbana degree won't.
Tier 3) Then you have the good but not excellent schools. I went to one of these for 2 years. These schools seem like slightly harder high school. You learn the material but are not stretched as much. If you didn't see it on a homework or quiz it won't be on an exam. 5 or 10 years down the road will it matter if you went to a Tier 2) or Tier 3) all things else equal? Not really. But your MIT or Stanford degree will.
1) Look for a job/project you want to do 2) Lie and claim you can do it, and commit to doing it 3) Learn the hard way how to do it. Because you committed to doing it, you can't quit when you get stuck and hate it. 4) Do just a good enough job to impress the people that asked you to do it 5) Do another project, this time doing it the right way (or at least better). 6) Repeat until virtually no one knows much more than you on the topic. 7) Profit!
I know inkjet still beats out laser for photos, but for everything else GO WITH A LASER. They're cheap, cheap to print, go long periods of time without a refill, print indelibly (vs. smudge-when-wet inkjet), and the ink doesn't jam up.
With (st)inkjet I found I was going through cartidges every 3 months whether I printed much or not at all. I think the ink must evaporate. Then I had a clogged head.
HP is of course the worst in terms of squeezing every drop out of you - printers that refuse to use non-empty cartridges after x prints and the like.
Slashdot Mirror
- AMD was on top of the world with Opteron / AMD64
- Intel was losing everywhere it went. You'd be hard-pressed to find an Internet / financial shop *not* buying AMD
- But Intel responded with Merom / Core2Duo. That mostly closed the gap, though initially the memory subsystem was still inferior
- Had AMD met expectations with the follow-on part (Bulldozer), there is no reason they could not have continued to win
- But in my mind, their ATi acquisition initiated their downfall. They became schizophrenic.
To beat Intel (like most market leaders) you have to have a non-trivial advantage. When AMD had one, they kicked Intel's ass to the point that they severely altered Intel's roadmap. When they no longer had one, they lost.
Last time I looked at buying SF property potentially to rent out I was scared off by the regulations the city imposes on landlords. How can anyone sleep at night knowing your tenants are more in control than you are? I'm going to bet I'm not the only one with this impression, which would mean given a choice individual owners will sell long before they'll risk getting into the 'affordable rent' game. Same dynamic may be why the developments are condos, not rentals.
So why no mention of the city's propensity to make being a landlord miserable?
For web applications, you're dreaming if you think MS is the most widely used platform.
Your post summarizes beautifully the fatal flaw with MS tools and Market-tectures.
They work great if you go down MS's beaten path.
They leave you in a lurch if you need to go off the path.
All along, Microsoft will trumpet the technology as the answer to everything and cause countless developers pain as they follow the same unreliable path only to fnid they can't ship anything interesting. It was a big eye-opener to me the day I realized that Microsoft told everyone to use MFC but shipped no MFC-based apps themselves.
Got to admit, though, that C# is pretty nice as a language.
I'm sure there are some brilliant nurses. I'm sure nurses are responsible for countless saved lives. I'm sure doctors could learn a lot if they spent time doing nursing. I'm sure that some nurses know more about practical human health mechanisms than some doctors do. But I have never heard of a doctor that got his M.D., then did nursing, then became a doctor.
I ride a bike 4-5x a week to work. My car often goes weeks - maybe a month - between drives. And I encourage everyone to do the same if they can.
But not because of CO2. CO2 is a trace gas comprising only .04% of the atmosphere. Humans emit only a small portion of the world's CO2, but it seems we must believe that the Earth can absorb and produce finite, unchanging amounts of CO2 such that any perturbation is disastrous. The climate data is not consistent, nor is the science predictive. Yes, pollution - as in things that dirty other things - is abhorrent. Yes, saving energy and money is great and we all need more exercise. But the CO2 obsession is a cult that insists everyone join.
...that the creator of C++ runs Windows. C++ is a barely-stomach-able OO implementation, and Windows is a barely-usable GUI & OS. I look forward to using neither, but often must do so.
A calculator without RPN is like a computer that only runs Windows.
Microsoft's problem is with software more than hardware. The Windows phones may not have been as slick as an iPhone, but the hardware sufficed in terms of durability, phone quality and screen. What did not suffice was/is the software, and I believe that is why Apple walked away with the smart phone market despite Microsoft's much earlier presence. Microsoft has existed in an environment for 30+ years in which poor or barely-passable software could lead to tremendous success. Those days are THANKFULLY over. Do they realize this? Or, just as Ringo Star claimed to be the world's best drummer (by association to the worlds most prevalent band) do Microsoft managers walk around confidently trumpeting that Windows is the greatest OS in the world?
The day that I can buy a non-Mac laptop that has a slick, polished, Unix-ish OS in a quiet, pleasing metal form factor is the day I even think about caring about cheaper.
Please do not use the word 'baddies'.
My job is to code. I had a neighbor that worked in a paint factory. He got hurt, and went to school to learn software. Pretty soon, he's coming home talking to me about UML / Booch and other esoteric OOP topics that I barely had a grasp of. I thought, gee, wow, it is amazing to see how fast a blue collar guy can come up to speed. But then it all kind of fizzled out. And I wondered for a long time why that was. How could someone get so far and be able to recite complex comp-sci topics and then give up?
And the answer is, there are (at least) 2 components to being a s/w developer. You need to be able to think and analyze and understand. But you also need to be able to - and more importantly - slog it out to make it work. And it's that slogging out that - at least in my experience - is where you truly grok the information and become effective and confident and useful.
So, that's why I say 'beware'. You might be having one of those 'I know Kung Fu' moments, because you can understand your textbook. And that is a wonderful thing. But unless you cement that knowledge by actually using it - and in particular, by learning to debug hairy problems - it is at risk of floating away.
I mean, I could pick up a cutting edge cardiology book and, given a few weeks, recite and even talk intelligently about topics that most doctors wouldn't know. I could probably impress virtually all of the population with this knowledge. But, I would be a long way from being someone anyone would hire to cut someone open. Balance your book knowledge with practical knowledge. You need a residency.
Holy Grail OR Delusion? A Holy Grail *is* a delusion!
Maybe it will be useless. Maybe you'll switch careers. Maybe you'd make more money working now than the increased pay later. Heck, in my case, I don't even have a bachelor's in CS and I do pretty well.
On the other hand, after a certain age it gets a lot harder to go back. Your time is suddenly spoken for by family, career, etc. Why take 3-4 years to finish part time at 2012 tuitions when you can knock it out in 1-2 years now?
In your posting you talked about adding ducts to steal A/C from a second unit. To work decently you would need to not only add a new output duct 'run', but also a new return 'run' (that is, unless the 2 units share a network of return ducting).
I remember how exciting the touch screens seemed in the late 80s, but when using them reality set in - you quickly fatigue holding your arm up to touch a screen.
Plus, with any user interface people need a certain confidence in correspondence between what they do and what happens. When you push a button, you KNOW it got pressed. If you push a joystick left, you KNOW you're going left. That 'payoff' is like a contract between you and the machine that goes favorably. But if pressing the screen where you believe you need to press may or may not do what you want, that contract gets shaky. Especially since there's no click or motion to reinforce what you're doing. This, by the way, is why I think 'free space' VR controllers never caught on...at least until the WII.
Still, software can create cues to take the place of physicality and have 'grease' to avoid common miscues. Plus, having the screen be horizontal reduces the fatigue.
But in the end, as archaic as the keyboard seems compared to touch and speech, it really is an incredibly expressive and low-energy-requirement device.
Now that I've been out of school for (gasp) 17 years I see it this way.
Before I talk engineering, I want to say this: the most valuable class for me in all my college career was the one that taught me how to write. A so-so engineer with excellent communication skills can go a lot further than the reverse.
Anyway...at one time because of demand for software people you really could squeak by without a very good degree or any degree at all. That's probably still true to some lesser extent. And I have to say that new grads fail to impress me with basic software *engineering* (debugging, organization, maintainability) skills.
But I *still* highly recommend you get the best degree you can get.
For engineering, my take is that there are like 4 different tiers of schools.
Tier 1) A degree from MIT and to a lesser extent Stanford is a gift that keeps on giving. For the rest of your career you will be referred to as 'that guy who went to MIT'. Even if you barely passed. Just like Gordon Freeman. Out on the west coast having a Stanford degree is about on par. But MIT seems to have students actually build stuff, which is not only fun but which also translates into people with better basic engineering skills. So, if you can manage to get in and graduate you will never regret it.
Tier 2) Then you have the rest of the 'top 10' or 'top 15' ranked schools. I'm out of date, but I'm thinking like Berkeley, RPI, WPI, Univ of Illinois Urbana, Michigan, some (but certainly not all!) of the ivy league schools and others my ignorance causes me to miss. Schools in this league are not only really helpful in getting your first job but more importantly they will WORK YOU. I went to what I call a 'tier 3)' school first for 2 years before going to a 'tier 2)' school, and boy was it a different level of intensity. That gives you skills and confidence.
The funny thing is that sometimes these schools rank above or darn close to the 'Tier 1)' schools. I went to Urbana, and I know there were years where it was #2 or #1. And from a curriculum standpoint you probably are getting just as good of an education. But - regardless of what your guidance counselor or parents tell you - even if MIT was consistently ranked #9, having that MIT degree buys you something that the Urbana degree won't.
Tier 3) Then you have the good but not excellent schools. I went to one of these for 2 years. These schools seem like slightly harder high school. You learn the material but are not stretched as much. If you didn't see it on a homework or quiz it won't be on an exam. 5 or 10 years down the road will it matter if you went to a Tier 2) or Tier 3) all things else equal? Not really. But your MIT or Stanford degree will.
Tier 4) Party school. 'Nuff Said.
You won't live to see it.
(music starts playing...)
Sunrise...sunset. Sunrise...sunset...
BLiPS - Bash Lines Per Second
Funny how God is the last explanation anyone is willing to entertain regardless of how much a stretch the alternative is.
You learn it the way any programmer learns it.
1) Look for a job/project you want to do
2) Lie and claim you can do it, and commit to doing it
3) Learn the hard way how to do it. Because you committed to doing it, you can't quit when you get stuck and hate it.
4) Do just a good enough job to impress the people that asked you to do it
5) Do another project, this time doing it the right way (or at least better).
6) Repeat until virtually no one knows much more than you on the topic.
7) Profit!
The people that write and/or know great code don't have the time to waste posting about it here.
Even with my 30" monitor, I strictly stick with 80 columns.
1) It's easier for humans to read reasonably narrow columns of text. That's why newspaper articles ('columns' if you will) and Blogs are narrow too.
2) It enables you to have many 80 column source windows open at once
3) It really helps out when others look at your code
4) It make life easier when you're stuck in front of a Linux console trying to fix code in a freezing datacenter.
I know inkjet still beats out laser for photos, but for everything else GO WITH A LASER. They're cheap, cheap to print, go long periods of time without a refill, print indelibly (vs. smudge-when-wet inkjet), and the ink doesn't jam up.
With (st)inkjet I found I was going through cartidges every 3 months whether I printed much or not at all. I think the ink must evaporate. Then I had a clogged head.
HP is of course the worst in terms of squeezing every drop out of you - printers that refuse to use non-empty cartridges after x prints and the like.