Slashdot Mirror
Review:Rise & Resurrection of the American Programmer
Decline and Fall
In the beginning of this decade, Edward Yourdon wrote Decline and Fall of the American Programmer, a pessimistic assessment of the American software industry in the global marketplace. I haven't read that book, but Ed conveniently summarizes it in the first chapter of this book.
His premise was simple: North American programmers are more expensive, less productive, and produce lower quality software than programmers elsewhere in the world. Therefore, he argued, competitive forces will drive them into extinction. He backed this up with a battery of data, figures, case studies, and anecdotal evidence.
This was apparently a wake-up call. After all, Ed wasn't merely an ignorant consultant. Rather, with 35 years in the industry, he had programmed mainframes and helped to invent structure design methodology. People listened to him.
Rise and Resurrection
But something happened during the first half of this decade. Not all of Ed's predictions came true; not all software development migrated to Bangalore, India; not all American programmers fell off the evolutionary ladder.
This book, Rise and Resurrection of the American Programmer, was written mid-decade to reexamine the situation then. Remember 1995? The web was just taking off, Microsoft Windows 95 was just released, Java was just beta, and Linux was just a hacker's toy.
Now the decade is closing, Ed is focusing on Y2K issues, and I've just finished reading this book. Some of Ed's views have dated, and some remain relevant. I'll try to enumerate a few of them here.
Java and the Internet
Ed stressed the importance of corporations embracing the internet. After all, he argued, competitors will. This has turned out to be correct. He also was extremely enthusiastic about Java, which wasn't even shipping when he wrote the book. Although most of the chapter explains the language and environment, and ends up sounding like a Sun white paper, one interesting nugget is Ed's suggestion that Microsoft may simply "embrace and assimilate" Java for themselves.
The Microsoft Paradigm
Let's be frank; Microsoft is quite successful, and does some things right. Ed dissects the corporation and comes to several conclusions. With section headings such as "The Dark Side of the Force" and "Into the Belly of the Beast," this chapter acknowledges public sentiment and should interest most Slashdot readers. But Ed concludes that Microsoft's hackers are growing up, and the corporation's powerful position will be difficult to assault.
Linux and Open Source Software
Ed pretty much missed this one. I didn't see any mention of Linux, hardly a blip on the radar when this book was written. Curiously, I also didn't see any mention of open source software, a wider concept that has been around for decades.
The closest hint I saw of the rising phenomenon was regarding Java's ability to change the economics of paying for software. Ed suggested that downloadable applets and web interaction make it possible to sell a "one-time usage" of software components, which could threaten existing software vendors with monolithic products.
An Enjoyable Read
At just over 300 pages, this book covers quite a few topics. I thought the chapters on peopleware and good-enough software were quite well done. Other chapters spurred me to learn more about the Capability Maturity Model and Personal Software Practices.
Ed is a frank and readable writer, and the book is quite digestible. It's fun to read recent predictions and analyze where they went wrong, and right (remember, hindsight's 20/20!). The book is almost a time capsule of the mid-nineties, which coincidentally being when I started working with computers in earnest was a refreshing read for myself. I think it will be for you as well.
Ed's web site is at www.yourdon.com.
Buy this book here.
TABLE OF CONTENTS
Preface
Trademark Acknowledgements
Part One: Decline & Fall Reexamined
1. The Original Premise
2. Peopleware
3. The Other Silver Bullets
Part Two: Repaving Cowpaths
4. System Dynamics
5. Personal Software Practices
6. Best Practices
7. Good-Enough Software
Part Three: The Brave New World
8. Service Systems
9. The Internet
10. Java and the New Programming Paradigm
11. The Microsoft Paradigm
12. Embedded Systems and Brave New Worlds
13. Past, Present, and Future
Appendix: An Updated Programmer's Bookshelf
Index
Write a best-selling book declaring doom, make lots of money.
Then, when it doesn't happen, write another best-selling book declaring joy and gladness.
Great strategy.
:-)
I've heard people say the same thing before. I can't see American programmers ever being out of work.
I think the book should be about how to succeed as a consultant/writer/pony-tailed "guru". Amazing that this guy could make money off of a book showing how his last book (which I'm sure netted him some green) was wrong. Simply amazing.
--
Jason Eric Pierce
make
make install
If these guys didn't have a language barrier and if they could afford modern computers to train on they might be a threat, but I'm not quaking in my boots yet.
He is making tons of money preaching doom and gloom about Y2K. Most of what he says on the subject is way off and almost sounds like it was written by Jon Katz.
Yourdon may have been a programmer 20 or 30 years ago, but lately his only experience has been "reading a lot of books". Maybe he does have a lot in common with Katz.
(writing this from the northeastern USA)
Back in the early 90's when I was considering programming as a career, I came across his "Decline" book in a bookstore and it made me pause for a moment and think "wait, do I really want to get into this field?"
Fortunately, I plodded ahead anyway (coz that's what I wanted to do) and I'm very happy now with my career choice. I felt a little vindicated when I saw the "Resurrection" book a few years later.
This technique has about 65% accuracy in weather prediction... much better than any of the computer models that have been developed. It also works well with other complex, chaotic, stochastic systems.
/. turned into an advertizing fourm... I was wondering when this would happen. Maybe I'll start posting Ads for used data communications equipment in the comments...
So... if I were an American programmer, I would start worrying.
Nice to see
Writing another book to sell that is basically a correction of his last book. You'd have to be a sucker to buy it.
>Let's be frank; Microsoft is quite successful, and does some things right.
So did Adolf Hitler!
I'm very surprised. If you can't even spell "demise" or "exaggerated", or can't be even bothered to perform a spell check, what kind of quality code are you producing? Now, I know that flaming people on the basis of their spelling / grammar is bad form, but I have met very few programmers that do spell things correctly, and this is a shame. Looks like nerds are indeed illiterate.
I think Lenord Nemoy (sp) wrote a book a long
;)
time ago "I am not Spock"
He told how he was nothing like the character he
played on Star Trek and how he hated being seen
as Spock everywhere he went.
A few years later he wrote a book called
"I am Spock"
Either he changed his mind or he had a degree
in business/marketing
Shouldn't it be the "Resurrection and Rise..." rather than the other way around? Kinda tough to rise if you're still dead...
Hi all,
:-)
I'm an Indian programmer, in US, loves and advocates Linux and GPL. However, there's been
quite a lot of flammable stuff about Indian programmer's here. Some points I've to disagree
with (on generalising opinions)->
1) They speak broken English.
One reason could be that they speak some
other language better. Still, there is
a reasonable quantum of programmers here
who are quiet proficient in English. Most
of the people educated in India know
multiple languages, some of which are very
tough. I am quite proficient in 6 languages
and can speak another one. Out of these 7,
two are foreign languages, viz., English
and French.
Oh yeah, I've seen many US citizens whose
English grammar falls slightly below awful.
2) They are poor programmers.
Literally so, may be. After all, most of
us come here, because India is poor and
we definitely earn much more here. Because
of this willingness and the demand which
encourage and exploit this, I agree (albeit
sadly) that a good proportion of Indian
programmers are adept at writing
unbelievably inferior code.
But this does not imply that all Indians
write bad code. Nor does it imply that all
US programmers write great code. In my
personal experience, the worst code I've
ever seen was written by 3 Americans.
I still can't understand why Slashdot could be
a forum for venting frustrations at the genetic
aspects of programmers. I doubt whether any one
of us who love Linux was aghast at Linus being
Finnish. Or did people start Gnome because
KDE was not started in US? Heh heh, do you
seriously believe that Windows is so popular
because Bill Gates is a US citizen ?
Point is if we are programmers, we don't need to
worry about the nationality or politics of the
authors. If we think code is good, learn from it
(I've learned whatever little I know, not from
text books, but by perusing GNU source code). If
we think it is bad, let us be graceful in
pointing out what is bad and how to make it
better (well, in case if something is hopelessly
bad, just say so
Another anonymous coward
PS : Some weeks back I remember seeing lot of
comments by one Natepuri. Now, please don't
use such mails to grade one nation's
temperament - or its people!
"The Road Ahead", you'd have to be an imbecile to buy that. I cannot believe that people actually think of Gates as a visionary and not a corporate weasel.
You should have told him to use a switch statement. The resultant compiled code would probably have used a look up table and the coder would have had the satisfaction of typing all the text. Everybody is happy!
// length // downstream freq
// length
// length
// length // length // length // length // length // length // 64000; chuck wants 0 for now. // length
// length // length of concatenation field // no concatenation support
// length // mic == 0 for now
// length
I'm working on a piece of code from *****(not MS), want to see an example:
*DataPtr++ = DOWNSTREAM_FREQ;
*DataPtr++ = 4;
*((unsigned long *) DataPtr) = (unsigned long) (DownstreamFreq * 1000000.0);
DataPtr += 4;
*DataPtr++ = UPSTREAM_CHANNEL_ID;
*DataPtr++ = 1;
*DataPtr++ = UpstreamChannelID;
*DataPtr++ = NETWORK_ACCESS;
*DataPtr++ = 1;
*DataPtr++ = 1;
*DataPtr++ = QOS;
*DataPtr++ = 28;
*DataPtr++ = CLASS_ID;
*DataPtr++ = 1;
*DataPtr++ = 1;
*DataPtr++ = MAX_DS_RATE;
*DataPtr++ = 4;
*((unsigned long *) DataPtr) = 10000000;
DataPtr += 4;
*DataPtr++ = MAX_US_RATE;
*DataPtr++ = 4;
*((unsigned long *) DataPtr) = 2000000;
DataPtr += 4;
*DataPtr++ = US_CHAN_PRIORITY;
*DataPtr++ = 1;
*DataPtr++ = 5;
*DataPtr++ = MIN_US_RATE;
*DataPtr++ = 4;
*((unsigned long *) DataPtr) = 0;
DataPtr += 4;
*DataPtr++ = MAX_US_BURST;
*DataPtr++ = 2;
*((unsigned short *) DataPtr) = 100;
DataPtr += 2;
*DataPtr++ = MODEM_CAPABILITIES;
*DataPtr++ = 3;
*DataPtr++ = CONCATENATION_SUPPORT;
*DataPtr++ = 1;
*DataPtr++ = 0;
*DataPtr++ = CM_MIC_CONFIG;
*DataPtr++ = 16;
for (i=0; i16; i++)
*DataPtr++ = 0;
*DataPtr++ = CMTS_MIC_CONFIG;
*DataPtr++ = 16;
for (i=0; iearn what I make dammit!!
Stupid stupid stupid. No attention to detail. You need to go back to school and get an education, sonny.
Well, bluGill?
"Jennifer Yourdon and I recently finished producing a video, in collaboration with James Talmage Stevens, entitled 'Year 2000 Home Preparation Guide'; it provides straightforward, common-sense guidelines for making prudent preparations for possible disruptions in the areas of heating, lighting, food, water, medicine, and other essentials."
If this video provides "common-sense guidelines" ... WHY WOULD ANYONE WANT IT?
-thomasI'm not American, but salaries here (England)
are much the same as in the states, so the
threat of hoardes of highly skilled, networked,
dirt cheap, 3rd world programmers reducing the salaries we can earn is just the same.
This guy writes a book saying: "we're all doomed"
and we do just fine.
Now he says: "er, actually everything's fine"
Doesn't that make anyone else feel a little nervous.
Programming has different qualities to most jobs.
A good programmer is 10-100x more productive than an average programmer. If you're really good, then you'll be fine whatever happens. If you aren't then I wouldn't be complacent about this.
I was in school recently, and thank god I didn't take a CS degree. Your computer science students have heard so often about how they're going to change the world; it's really started to fuck with their heads.
Anyway, my experience is that the students you decide are "better programmers" are the ones who talk more than the rest, and tend to blow smoke. I tend to find that they don't know what they are talking about, and are focused more an appearances than on computers; developing a cult of personality rather than developing programs.
I once listen to a classmate (in a first year class, no less!!) tell me how NT was more secure than UNIX, and how much he knew about these issues. I managed to hold out until I left, but shortly thereafter I laughed until I couldn't stand up. Egos at least provide comedic relief.
"I'm betting 5 years."
Hey, I've survived for 20 years, through Jimmy Carter's
stagflation, Reagan's recession, Bush's war, and Clinton's
affair. Sure, I'm now churning out C++ on a Linux PC as
opposed to Z80 asm. on a box with 256K of RAM,
but nothing much else has changed. And that includes the
useless "predictions" of pundits like Yourdon.
I haven't found this to be true at all.
"Statistics show only a small percentage of people who program when they leave
college still doing it 20 years on. "
Well, I'm still doing it after 20 years, but I think there are
several reasons why most don't:
1. Some move into management, where there's more money
and respect.
2. Some go into law or accounting, where their computer
experience gives them an advantage.
3. Some can't adapt to changing technologies.
4. Many find that they just don't have the talent for writing
software.
What these predictors of gloom-and-doom for us overpaid
Americans forget is that economics is not a zero-sum game.
When those Bangledesh programmers start getting $1000 a
month, that puts them in the top income scale where they
live. They can now afford to buy their own PC's, and software
to go with it, which increases the demand for programmers.
They get on the internet, see that programmers in Europe and
the US are getting paid way more than they are, and start
demanding comparable salaries for themselves. With their
increasing incomes, they'll want better housing, cars,
appliances, causing other parts of the economy to improve.
Eventually, Bangladesh will start looking (and costing)
more like Hong Kong or Singapore. The PHB's will have
to look elsewhere for cheap labor. I predict it will be
Mexico and S. America, and that Yourdon will have the
opportunity to write a couple more books about it.
I'm actually in the process of reading this book. It's tedious and dry.
Ed Yourdon has been around for a while, but he reminds me of an episode of 'Newsradio' where a visionary with great big ideas is interviewed on the radio. He announces with great gusto that computers are the future; and they can do anything. He doesn't have one yet, but he's going to get one very soon.
It amazes me that anyone pays attention to the man anymore. He writes a book saying how American programmers are doomed, then finds out that nobody wants to pay him big bucks to come and talk to them about how they are doomed, so he writes a new book explaining how great everything is going to be, since all the COBOL programmers (the only kind of programmers he knows about) are going to become, get this, webmasters! Then he writes a book explaining how civilization is going to collapse on 1-1-00(R) [trademark used without permission] because people's microwave ovens are going to quit working because they have computers in them, and all computers are going to fail on 1-1-00(R) [trademark used again without permission]. Give me a break!
Linux (Finland)
Amoebe-Minix (Netherlands)
Chorus (France)
Pascal (Austria)
C++ (Denmark)
I don't think there is a language barrier. After all, algorithms are universal. (like mathematics)
Regards,
Philippe
BS! Visit the web pages of the top CS & EECS
schools in the US and check out the percentage
of current or past MS and Phs students that are
not natives. Of course, you cannot always tell
from the name. In my old school (top 10 school),
it breaks down to 7 natives vs. 11 foreigners
that have graduated with a PhD during the last
year. Even with the foreign input, the schools
just don't produce enough people with the right
skills.
There are two reasons that really explain why
this happens. First, the public education system
just sucks in this country. People are debating
whether physics should come before chemistry.
We are talking about one year of physics and
one year of chemistry! In my country, everybody
has to take the same physics, chemistry and
math classes for the first five years of high school.
Only in the last year, people focus on
specific directions. The same joke occurs in
US colleges. To get an undergraduate degree,
students fool around for a couple of years
until they decide what they want to do. True,
motivated students can get a lot out of US
colleges but there are not nearly enough.
The second reason has to do with rewards.
Engineers might be well paid but nowhere close
to what lawyers or doctors get. And since dollars,
not knowledge, determine recognition, I am not
surprised that not many students choose this
path. It's not that other societies are not
materialistic or crude. But somehow they do
a better job educating people or at least
ehy used to do. The US educational system
was not bad fifty years ago either. Who knows?
Maybe, the whole world is going down the tube
in this aspect.
For all these reasons, I intend to get out
this country as soon as my kids have to go
to school. It's true I can find private
schools that dodo provide good education,
but there is something broken with a country
that makes adequate education available only
to the elite. My wife, who happens to be a
native, also agrees with this assesment.
BTW, I don't expect me to understand what I
am talking about. US is the world leader,
after all, right? So was Rome. Enjoy your
bread.
I immigrated from Middle-East to Canada as a teenager. 14 years after, I was still in Canada making a living and paying all my taxes, when a German who'd just moved there 2 years before told me to go back to my own country!
Another story: Sometimes when a "white" cuts me off in traffic, I roll down my window and shout at them "go back to your own country!" They always laugh. (This is Canada, I wouldn't dare doing it in US)
The important thing is that his code worked, and that it was easy for someone else to understand. Cleverness is highly overrated in programming, and clarity is highly underrated. You're not a better programmer just because you can write an advanced operation on one obfuscated line full of bangs and ampersands.
"but everyone is looking out fo their own "
^^^^^^^^^
define your own. does this include people who were born elsewhere but have lived nearly their entire lives here?
"...and when foreign programmers undercut our prices we get a little irritated."
that's capitalism. blaming the programmers misses the point. if you want someone to blame, try the management of the corporation which has elected to go with the cheaper option. (they're selling out "their own people".)
if you want to stop blaming, and do something useful, expand your skill set so that you have something to offer that the competition doesn't.
Look believe it or not it is important to write code this isn't optimaly slow. The code that the coder produced is difficult to maintain (try changing the substitution) and difficult to debug, for example:
if (character == 'w')
{ character = 'm';
} else
if (character == ('w')
{/* coder meant 'x' */..
character = 'n';
}
It is downright poor code. It's slow too. I know that prof's often say 'speed if over emphasized' but they say it so often that it is now underemphasized.
The moral, your professor teaches because he can't handle a job. Working pays a lot more than teaching at this time. A hell of a lot more. I make more money than any of my professors did, and I'm admittedly stupider than (some) of them.
I once wrote a hash table to handle IP addressing in a cable modem, they wouldn't let me use it because somebody else had solved the problem (we were a disorganized bunch) with a table that just searched until it found a match. This was on a 33 Mhz ARM processor with a 16 bit wide bus, and a table of up to 256 addresses. That was one of the reasons I finally gave up and quit. My table would statistically never go more than 2 deep into the search and I PROVED the hash algorithm was uniform, but the other solution would do an average of 128 deep. My solution was judged 'inferior' and people were suspicious of it even though I tested it with random table deletions and additions over a period of 2 days without an error with a random number generator as input (getting collisions of the hash table up to 5 deep, never more). Everybody was using methods I was using in highschool and had the education to boot.
It is really fustrating to work with people like this. They fuck up projects.
You're forgetting that you "can't get blood from a stone." You can study all you want but, if you ain't got the ability; you ain't going to comprehend abstract concepts (e.g. OOP).
regarding "Asians", they're "over-represented" in technical professions; which correlates well with the average asian's IQ being higher than the average caucasian's IQ.
Anyhow, what makes the U.S. resilient is its natural resources and its ability to attract the "best and brighest" from other countries (aka brain-drain).
is the worst sin.
1) You write code that works first, is clear,
and does what your specs require.
2) You reanalyze the requirements, and refactor
your code so that it is easier to be modified,
to meet your new requirements.
3) iterate these two steps for awhile
4) analyze your program for bottlenecks, which
will occur in 20% of your code. Optimize
those.
I mean, in the case of the substitution cipher.
You could easily start out with a switch statement
if you wanted.
Step 2 would be adding an API for generating and maintaining substitution tables.
Step 3 would add the ability to "plug in" new algorithms for performing the substitution using the "Command" Design Pattern, and so on.
Over time of course, you learn to do some things up front. But no productive programmer will ever do everything correct up front.
Those who attempt to design perfection up front, end up not meeting their requirements and finishing work on time. And they still end up having to refactor their code.
thats to their advantage.
in many schools people are taught that america
= everyone has a fair shot but in reality
theres alot more bulshit that has nothing to
do with how good you are but everything to do
with how much golf you play with your boss
and how much ass you can suck....
of course this is stupid and not the way things should be,
and if a country ever doesnt have that stuff then it will be at an ADVANTAGE
not a disadvantage.
If it is either nationality of origin or genetic characteristics, tons of programmers at Microsoft as well as many other software outfits are Indians (Pakistanis, you are included too).
I am sure these guys don't work in an island. So, in this highly connected world, what is it that Yourdon wants to say ? That the so-called "resurrection" (without the "demise" ever happening) owes it to those Indians over here ? Atleast that would have been consistent...if not an unfounded theory (again!). Gimme a break, Yourdon ! --A. Coward
Well, there's another one, assuming you meant statistical, and that yo're a different guy. I think he's basing it on his experience of programmers and /. posts in general
By which I mean,
/. posts in general."
"Well, there's another one, assuming you meant statistical, and that you're a different guy. I think he's basing it on his experience of programmers and
(1) Australia is a well-off Western country which relies heavily upon natural resources for its wealth.
(2) Given freedom to do so, people of course will travel to those places where they can get the best rewards for their skills.
For reasons of past peformance, there are a number of highly regarded US Universities. Graduating from one of these will have more weight often than from a local University, especially to employers in the US.
Also, with the demand for highly educated people in the US being so great, comparitively huge salaries are offered. So it's no wonder that there is an influx of bright visitors and immigrants to the US.
This has nothing to do with the current _general_ quality of education in the States.
The fact that a country with such a large population and generally high quality of living can't come close to meeting its own demand for skilled programmers tends to indicate that something is terribly wrong with the education system at large.
There has been a lot of talk about good coders vs. bad coders. When it comes to hiring & interviews what is assesed seems to be experience and the person themselves.
So employer wants someone to work on, say, a browser & Java based frontend for a DB on an AS/400. Someone has experience with Java applets & AS/400. They can't tell if the guy wrote crap that never worked, or if he developed a good solution.
Essentially there is no way (as hiring is currently done) to assess if someone can take a problem and and form a solution in a programming language, and understand what is doing etc. In college we did software classes, not everyone had any real understanding of what was going on, or could express solutions in code.
Experience does not neccesarily mean you were any good.
I was born in South Africa, hold German citizenship, live in Britian, employ programmers in Poland, South Africa, England, Germany. I am looking for people to employ in India, Malaysia, China because these are a good source of cheap labour, and the internet makes . My point is that where I purchase labour is driven by the same needs that drive where you shop for your groceries, and location is only a part of the equation (and a part that is declining in importance with the internet). Quality versus cost is the driver. (And by cost I mean all the costs, not just salary.) In my particular case I work in an industry with low margins - I can only make a profit if I keep costs down. Labour is the biggest single cost. Therefore I maximise for quality/cost within my spending limits. At the moment it's clear that for the most part most companies in the US feel that they maximise their return by buying local labour. I wouldn't expect this to be the case in the long run, in just the same way that it hasn't held true in any other sphere. Cars. Cameras. TV's. Once the US was the number one producer of all of these.
When I was at university here in the US, I remember one time our assignment was to write a simple substitution-cipher program. I came across one of the "better" programmers writing out code like this:
if 'a' then
if 'b' then
if 'c' then
And so on... And this was for [a-z] and [A-Z]!! I tried to tell him that this could be done in one line of code, but his response was that since he had already typed in so much, he wanted to use what he wrote. I find this really sad. It would have been quicker to write a program to generate the 52 lines of code than to type them in!! No wonder Microsoft code is bloated, if this is the type of programmers they hire!!
Being one, and having been trained with others, I can say that many of the newly graduating programmers are ill equipped. Too many people are taking up programming with an eye solely on money. That's all I ever heard classmates talk about before class... "I'm gonna make 60K / I'm gonna make 100K / This guy I know makes $x / etc." They learn enough to pass their classes (make a small database and a card game, or something), and get hired by people who know less than them. Apathy does not make good code, nor does an eye on your next raise.
Obviously, there are some damn good American programmers, but the rest of us could be better. I imagine that anyone reading this is pretty good (or at least trying), as it shows the personal initiative of reading information on your field, but how many others just toe the line for a check? If those "programmers" end up losing out to better foreign programmers, that's their own fault. Would you rather have the air traffic control software for your flight be programmed by a shoddy textbook American or a foreigner who takes great pride in his work? Regardless of nationality, we all need to constantly assess ourselves. It's our skills that make us valuable, not our nationality.
This guy picked up on the y2k hysteria bandwagon and lost what respect I had for him. Blech.
I talked to a computer science researcher at the university of Kyto last summer, and he explained something very interesting about the way the language barrier worked:
Of course, you have to know english to be able to do comp sci. It's what the vast majority of programming languages are based on (I've yet to see otherwise, in fact). Understanding english-based programming languages is not a problem for the japanese researchers. The problem is generating new ideas, and expressing them in english. If, for example, they want to create a new language, it has to be in english if they expect anyone outside of japan to use it. However, not being native english speakers they cannot conceptualize and express it effectivly, and this really hurts their position in comp sci reseach generally.
Japan's main software industry is in entertainment. Otherwise, they use the Japanese version of Windows, the Japanese version of MS Word, etc... Software is imported and translated, rather than created at home. Basically, the American programmer is in no danger simply because of a natural language advantage, and an already dominating position in the global marketplace.
Recent events made him change his mind, eh? Like the events the proved his original thesis was completely wrong?
What you're missing is that many of these Indian programmers are Americans! A very large number of Indian programmers are US citizens or wish to become US citizens.
Well, America was built on immigration. A lot of the "foreign" programmers that come here do so with the hopes of becoming a US citizen or getting permanent residence. (Some are undoubtedly here just to make some cash as well). The fact that they can't get those types of visas easily is not necessarily their fault. So a lot of these people should count as Americans, IMO. But I agree with you that American companies do value the skills these people bring.
Predictions of massive loss of American jobs to foreign programmers reminds me of people constantly saying how America's poor educational system was going to cause us to fall behind competitors in Japan and Europe, whose students dramatically outperform US students in international comparisons. But guess what, the US is holding its own just find with Europe and Japan, despite an entire generating of people growing up in our supposedly third rate education system.
What I think these analyses fail to take into account is how American culture makes up for a lot of things. Americans have a more anti-establishment, anti-authoritarian outlook than people in most other countries. I think this tends to lead to more questioning of the current ways of doing things and thus a lot of innovation (Java being a great example from this book). Americans have certainly been heavily involved in most major software innovations. Additionally, America is very socially mobile, and has a culture that encourages people to take risks by starting busiesses, or jumping to a startup in the hopes of long hours and quick bucks. America is also fixated on freedom and rights so it should surprise no one that the "free software" movement had much of its genesis in America. In short, people in some other countries might be able to reguritate facts on an exam better than we can, but we've got a few qualities of our own that can make up (in some cases more than make up) for that.
The above paragraph is not meant to suggest that other countries are filled with sheep satisfied with their lot in life and with no ambition or mobility. I'm only saying that the elements I highlighted are more prevalent in America than elsewhere (which is something that many promient scholars agree with, BTW).
Posted by The Mongolian Barbecue:
This is yet another example of how most pundits and generalists are full of shit. They make a career out of pulling ideas out of their asses. If they guess right about things, they claim to be prophetic, and if they are wrong, they make a bunch of money updateting their crap
Of course there are good programers from India. I've met some.
There are a couple problems they have not overcome though. One is language. I don't expect them to be good, but they should be at least passable. According to the mythical man-month writing code is 1/6th of a programing effort. My expirence backs that up. So being good at writing code doesn't help these excelent programs (who are not particularry better then Americans) do the other 5/6th of the job. Still overall I can deal with a few of these programers on the team, I'll make up the slack, they do add to a team when there aren't so many that the other 5/6th goes lacking.
The porblem is my expirence is that these people are a minority. I worked with them in College, since many of the other students were not American. Most were not compentant programers. They couldn't be trusted to write the code that needed to be written in such a manor that I could go in after them and make a minor change. Unmaintainable code is a death toll in the real world. Much of the Y2K problem exists in code written in the 50s and early 60s! Mind you I've seen americans who write code just as bad, and just as many. The difference is I can shunt those americans off to the other 5/6th of the task.
The sterotypes exist for a reason. Some sterotypes exist for the wrong reason, others are no longer true, but most sterotypes once existed in truth.
Yeah, but the ones that are here earning $30/hr aren't as big a threat to my livelihood as the ones in Bangalore earning $30/month. I welcome competition against the foreigners who are here, because they have to earn a US standard of living same as me.
As a Canadian working in the US (thank you NAFTA), I can't believe I just said that.
The next Cmdr Taco duplicate will be ready soon, but subscribers can beat the rush and see it early!
Remember
The ties may bind...
... but if they are made of leather they also chafe
Well, I can see where you're coming from, but...."If not all your predictions come true, then never write again." Is that what you're saying? Who's gonna tell Arthur C. Clarke? :-)
dylan_-
--
Igor Presnyakov stole my hat
Well, Yourdon is always right. Actually he says that in his book, with examples. Prediction is a hard thing.
A lot of the book isn't about prediction. It's about bettering yourself and your company. For example, by using best practices, or reading books on your own if your company won't train you.
The book is a light and interesting read, even if it gets somethings wrong. Even in the opinion section, I don't agree with it all. And it is several years old, so has more historic than practical value right now.
Personally, having only entered this industry in the mid-nineties, I missed the whole IBM/COBOL/decline thing in favour of the Linux/Java/resurrection thing. And I'm not even American.
Actually, the more I think about it I'd give it a 7/10, simply because I don't think it rates as highly as More Effective C++ and wish to be internally consistent with my ratings.
--
Marc A. Lepage
Software Developer
He has written at least two other books since Rise and Resurrection. He also coauthored another book in the same year.
See http://www.yourdon.com/articles/articlesummary.htm l#Technical books for more details.
Does this mean that there will be a civilisation after 2,000, and that angry mobs won't string all programmers up by their entrails in the centres of looted, burned-out cities?
Sorry I'm late.
Amazon $13.56
BarnesAndNoble $13.56
Shopping $11.01
Spree $13.56
1BookStreet $15.26 (includes shipping)
Regards, Ralph.
You guys are way to harsh. Yourdon's first book was a wake up call to the industry, and anyone who was around at the time knows it. The fact that many of his predictions didn't come true reinforces the idea that his first book had good value as a cautionary tale. It got many people to change the way they looked at their careers, to get off their butts and do something before the worst case scenario came true.
Just because you have a bad case of 'pony tail guru' envy doesn't mean you can't do something about it. No one is stopping you from getting out there and writing the next big gloom-and-doom best seller. Go for it!
"The only good windmill is a tilted windmill."
Not to mention Linux, Apache, etc...
"The only good windmill is a tilted windmill."
"Writing another book to sell that is basically a correction of his last book. You'd have to be a sucker to buy it."
Why does that make me think of Windows98?
I see even classic Slashdot is now pretty much unusable on dial up anymore.
If Amazon wants to subsidise my slashdot habit, great, but isn't this a review of a book that's been out for three or four years? And isn't it about time for a new one from him where he says this one and the one before that were all wrong?
I see even classic Slashdot is now pretty much unusable on dial up anymore.
Just a personal thing, maybe. I used to subscribe to Computer Language magazine back in the late '80s, and enjoyed it very much. Then Yourdon began writing a column for them, and the editors seemed to place great stock in his prescriptions. I found them to be tedious, annoying and just plain wrong. Consequently, I discontinued my subscription.
Do they still publish, now?
Peace and love, y'all
You're talking about stereotypes? You basically just said that Indians and SE-Asians in general are better programmers because they're "foreigners". I think this assumption is based on the fact that are a substantial population of them in University but this has more to do with the fact that it's the "thing to do" - not because they are inherently better at it than your average "white folk."
Remember the 1990-1991 recession? Laid-off techies were moving out of Silicon Valley in droves, so many that the moving van companies couldn't cope.
Those of you in your twenties don't remember hard times. They will come again. The business cycle hasn't been repealed.
If you're tops in your field, you can do well even in recessions. If you aren't, or you were tops in your field but are getting a bit rusty, or you manage to get into a specialty that becomes obsolete, you may have big problems.
Comment removed based on user account deletion
The American programmers who leave after 5 years are generally people who have gotten tired of it. A lot of people who get teaching degrees burn out in the same fashion - they find it's not what they want to do with the rest of their lives, or it's not what they expected it would be when they were studying to get there.
Don't confuse this with people who can't get jobs.
Also, when talking about shipping programming jobs overseas, bear in mind the difficulty of coordinating that work with the programmers back home. This is definitely not trivial, when the time zones make it impossible to even teleconference during business hours.
Jon
All opinions expressed herein are my own, and not those of my employers, who are appalled.
Stupid patriotism.
Are we really waiting for such a book?
Maybe yes, but why the **** AMERICAN!?!
or is it just me?
Try _Death March_, by the same author. It's like _The Mythical Man Month_ but even more blunt.
I've finally had it: until slashdot gets article moderation, I am not coming back.
Yourdon has pretty much been discredited in the software engineering community. The irony of this is that "Decline and Fall" was 50% right! In fact, there *are* many software engineering jobs moving to places like India. I know of several American companies that have opened operations there, or purchased pre-existing Indian subsidiaries.
The difference is that high-tech jobs have grown at a much higher rate in the US, and despite the fact that the Internet makes the US->India system possible, there is still a lot of value in physical location (e.g. Silicon Valley/Alley/Route 128).
One problem with many of these "sky-is-falling" types of predictions is that while they argue vociferously FOR their prediction, they fail to consider the another alternative -- that both systems prosper.
He did it again. His latest book is a doom and gloom Y2K book. Heard all about it on Art Bell, your quintessential fountain of paranoid delusional ramblings.
by Mike Buddha -- Someday the mountain might get him, but the law never will.
I thought many American programmer were already out of work. Statistics show only a small percentage of people who program when they leave college still doing it 20 years on.
A lot of work certainly has moved abroad, and cheap labour is imported. I've spent quite a bit of time in Bangalore over the past year. Every street has some well known western companies, and each is employing 500 to 1000 people there. Nobody seems to bother with a small operation. It only makes sense to go off-shore if you do it big time.
The results people get with these off-shore operations is very variable. Some of the best stuff in the world comes out of Bangalore - mostly software, but some chip design too. Some really low quality, low productivity work is also done there. Perhaps the results people get out of these operations reflect the effort they put into them.
Now Hydrabad is on the rise, with thousands of prgrammers doing mundane Y2K work there. Chip design is also building up in Bangalore, Madras and Hydrabad.
In Bangalore a reasonable programmer gets US$500 to US$800 a month. Even if they were not as productive as some other countries they would still offer could value.
Cheap well educated people who speak good English sounds like an investor's dream come true. Perhaps the migration to India will gather pace now it has been seen to be successful. On the other hand it may be capped by a limited of suitable people. Prices have certainly soared in Bangalore in the past five years.
I was particularly frightened by the monetary figures... $500-$800 a month? Wow. I work for a VERY LARGE telecom as a contract consultant (programming, Oracle, etc) and they pay about $500 A DAY for me to be here. My consulting firm gets a cut of that, of course. So moneywise, it's certainly prudent for companies to look overseas. But...
I work with a variety of people here. Indian, Romanian, Hungarian, and even a few Americans. My software group is 7 non-americans, 2 americans. But I don't worry. You know why? Because even though these people are smart, and write good code, they just don't get the big picture problems. They don't troubleshoot very well, and they don't self-task very well. They don't manage other people very well. They can't get their mind around the large issues, they don't understand the American way of doing business and serving customers, and they don't understand corporate politics and schmoozing. Their communications skills aren't as polished. Certainly there are exceptions, but that's my overall impression after working for 4 years in this industry. Also, I'm not defending some of the stupid corporate crap I put up with, but hey, that's what pays my (high) salary. I take the good with the bad. But I don't believe, not for a minute, that the majority of these programmers could do my job. If we pared my group down, one by one, I'd be the last to go. Multiple skillsets, superb communications skills, and cultural knowledge of how we do business are truly valued assets in a programmer. Why? Because so many programmers don't have them.
Don't worry, be happy. And stash large amounts of $$$ for the future if I'm wrong.
Why are we even discussing this?
I owe a lot to Ed Yourdon for his one truly great book: Managing the System Life Cycle. I unreservedly recommend it still, and wish he would update it.
Decline and Rise both had some interesting history and surveys of the big iron software world, and I think his emphasis on software quality, SEI and the like are good, but frankly he's missed out on all the important new things happening since 1980, including but not limited to freeware/open source.
Rise was published in 1996 and was out of touch even then. Why are we even bothering?
--------
Bill Gates Is My Evil Twin.
I dont see a need to distinguish between American and ,say, Finish programmers. When you become a programmer, you enter a culture that trancends the globe. You might be a citizen of country X or country Y, but if you program, your culture is one culture. Why do you think Linux is so good, because this culture has come together in one of its first collective, cross national efforts. Linux demonstrates the strength of these ties, despite great distances, and language barriers. Lets face it, when you program, you speak a common language. Granted, you might speak a dilect like C or Fortran, or whatever, but the roots of the language are the same. Programming is the tie that binds :)
-Master Switch, one more element in the machine
Well, I've tried to hire qualified programmers, and it's a difficult thing to do. And it's not because they're being paid such poor wages -- I'm a programmer and I'm making a very good amount of money. But maybe it's just that I came here from the education profession. Maybe I just don't realize how much more deserving programmers are, but as a schoolteacher, I worked harder and was paid half as much.
Some of the Redmond-ites might see the place as a concentration camp, but Bill isn't exterminating them. In fact, they hang around till their stock options vest, and leave the place with a cool million or three.
Comparing Gates to Hitler is just plain stupid, and shows a lack of imagination.
Eagles may soar, but weasels don't get sucked into jet engines.