I work as an architect for one of the largest companies in the world. There have been several attempts at finding the standard language holy grail, and in the end, the strategy that works, is to have a standard language assigned to given problem domains. With that in mind...
* The web site itself is done in Java/JSP/JSTL/etc. * The business logic is built with PL/SQL * The scripts that build the site are written in shell, make, m4, Perl * The QA team gets along marvelously with Python * The end-user, on-demand digital goods delivery offerings are written
in either Flash or.Net, depending on problem sub-domain * Etc.
The policy is to find what's the best tool for a given job, and standardize on that tool. The strategy has worked very well.
My suggestion? The good old "find the right tool for doing a specific job".
I manage a small but important team. The guys who report to me are, by the definition of their jobs, highly technical. Whenever something complicated needs to be researched and/or implemented, my guys get to do it, especially if it has to do with the adoption of new technologies.
We had our quarterly review a few weeks ago (it goes both ways; they evaluate me, I evaluate them) and the results were excellent. Here are the overall management techniques I employed with them:
1. Hold everyone in the team, including myself, to the highest
standard.
2. Define what 'highest standard' means as a part of the requirements
specification.
3. Once a decision has been made, by the team, business owners, etc.
there is no arguing. Part of my job is to keep the business guys
from becoming a distraction. The other part is to ensure that
the engineers deliver (1) and (2).
4. Go through a quarterly review with them; divide a sheet of paper
in three columns labeled as follows:
a) Desired outcomes (projects, training, coaching of others, etc.)
b) Achievements
c) Areas that need improvement
At the beginning of the quarter first quarter ever that you
implement this, fill-in only items in the first column. At the
end of the quarter, fill in the other two columns. A person is
doing great if they had, say, four desired outcomes and wind up
with four or more achievements. Last, review things that need
improvement (mine is "needs to attend relevant meetings" for this
quarter). Discuss those AND FOCUS ON BEHAVIOUR, not on
personality. Explain why the improvement is needed. After you
negotiate what this means, add it both as a thing to improve and
as a desired outcome for the next quarter. Repeat every quarter.
5. Respect your engineers' decisions. Combined, they know more than
you do, regardless of how technically capable you are. If that's
not the case, you shouldn't be a manager and you're probably not
meeting 1-3.
6. Leave your engineers alone to do what they do best. Don't invite
them to too many meetings or have them do tasks unrelated to their
charter. Engineers hate distractions, and distractions prevent
the team from achieving 1.
7. If the business folks start coming up with eleventh hour changes,
ensure that the engineers are part of the discussion and reason
WITH BOTH SIDES to figure out which changes make sense and why, which
don't, and how to come up with a solution that will meet everyone's
goals. NEVER just inform the engineers that a decision that affects
what they've been working on for three months has been made.
8. As a part of 4, create an environment where you are constantly training
your team, exposing them to new technologies, etc. Reward the intro-
duction of new techniques, procedures, etc. In 4, suggest that they
read at least a new book ON SOMETHING NEW NOT USED AT WORK every quarter.
If you work in a Java shop, they should be reading about Ruby or.Net.
You never know when a better mousetrap is available if you aren't informed.
9. Reward excellence whenever you see it, from solving the thorniest algorith-
mic q
As someone who's been programming in Java for almost 10 years and developed several GUI applications and such... yes, I feel like these guys shouldn't have used non-standard stuff that doesn't work out of the box. I don't care if JGoddies is open-source or otherwise. All I care about, as a follow-up poster noted, is that I tried several standard ways of running this in Java and it didn't work. And I blame that JGoodies library because Columba still doesn't work even if JGoodies is in the -cp/CLASSPATH.
As an end-user, I don't care if it takes ten.jar files and a prayer to make this run. It didn't work for me and it'll be faster for me to install something else (i.e. Thunderbird) than trying to figure out why this failed.
Maybe it's just a packaging issue but I find it to be a huge turn off either way. Since it relies on open-source libraries, and they ship them with the code, they could at least ship a.sh/.bat file with it that does all the class path magic. A README with instructions on how to run this might help. An installer or a self-contained package would be even better (download and install Thunderbird for a good example of this).
I just downloaded and tried to configure Columba 1.0 under OS X 10.4.2. My verdict? Skip it.
The people behind Columba used some widget library that's system dependent. This is throwing a number of null pointer exceptions under OS X with the Java 5 JVM. They all relate to something called "jgoodies"; they're doing something that appears to be system dependent.
One of the main reasons for using this would be portability. They seemed to have missed the boat altogether since it doesn't run under an otherwise standad Java configuration! Why bother with writing a Java application if it's not cross-platform? Why use non-standard widget libraries? Attaining cross-portability in Java is hard enough as it is; these guys chose to make it even harder. Thank you for blowing away the only reason I might've had for using the Columba email client.
You can see a screen capture showing the exceptions here:
Can't say if this works at all because I was unable to tell Columba about my IMAP server. I got another of those jgoodies-related exceptions when I tried to select something other than POP3.
A few years ago I learned a trick from my local Swatch store. I had scratched the face of a watch given to me as a gift; a keeper mainly for its sentimental value. The face had several scratches, some looked deep. I steeled myself to pay $50 to replace the watch's face (an $80 watch) and headed to the store.
The clerk was very helpful and passed on one of the best tips ever: Put some toothpaste on the polycarbonate surface, rub softly with your fingers, and wipe off with a moist cotton pad or paper towel. Scratches be gone! I've used that trick on mobile phone screens as well, with excellent results.
I can't remember the name of the painting off the top of my head, but that's a mink, not a ferret.
The symbolism of the painting is that the woman is so pure that a mink, usually very, very, very fastidious about the cleanliness of its fur, lets her not only touch but even hold it. These little critters don't like *anything* touching them.
If I remember the name of the painting in the morning I'll come back and post a follow-up.
You may want to take a look at MontaVista's real-time Linux offerings (http://www.mvista.com./ I was the technical lead for a real-time Java controller core that ran atop Hard Hat Linux and implemented both BACNet and Profibus (via Applicom dedicated I/O cards) for building automation and industrial robots, respectively.
I sold the rights to the software to the oil drilling equipment company that implemented the industrial robots but I'll be happy to assist if you want to discuss. You can find me as pr3d4t0r in irc.freenode.net ##java or chupacabra in Undernet #java.
I'll be a speaker at the Java in Action conference in Orlando this coming October; one of the sessions will talk about recent work I made in embedded/mobile/full-automation stuff. Most of my work is now based on Linux with some OS X, Solaris, and QNX to spice things up (I now work full-time for someone else; got tired of the startup game...)
Seriously, I haven't used a "word processor" in years, except to read stupid infected files other people send me. Spreadsheets are more handy, but Appleworks/OO.o do just fine in this area.
...and some people wonder why they can't find a job...
Interesting comment. For the record, I came here as an H1-B in 1990, before there was too much demand (i.e. I got my green card in less than 18 months). I never had trouble finding work or otherwise keeping busy. The same applies to a few of my friends. And that's my point. It's not about finding the top 1% of anything. It's about finding people who are qualified to tackle a job. And a lot of the people I get to interview I wouldn't hire.
The reality of things is that businesses nowadays have to strive to operate at that top 0.01% or someone else will eat your lunch. It's not pretty, it's not easy, but it's factual. If you don't put 100% into what you're doing, someone else will come and knock you off the perch.
The business that my company is in the most competitive in the world. Unless everyone is pulling in the same direction AND qualified to do their job above mediocrity levels, someone else will come and take our place... and our jobs.
The difference between being a professional and an amateur is in the qualifications. I don't want someone with paper credentials. I want someone who can deliver results for herself and for the company. I don't want the top 0.01%, but I don't want anybody in below the 65% mark either.
(FYI - 65%, no curves, was the passing grade at my university. Anything lower than that was an F. That helped to keep us on our toes.)
I'll think a bit more about your post; I'm a bit too tired to answer more coherently and my answer may not do justice to your comment, which I think it's insightful and will help me in future decisionmaking.
I work as a systems architect for a very large company in Silicon Valley. The problem is not that there are a large number of unemployed programmers out there. There are. The real problem is the lack of qualified programmers if you want to do something serious.
The company I'm working for has about 100 engineers in various capacities. Nobody there has less than five years of experience, nor can we hire anybody with less than that for the work at hand. There are a number of positions open because the QUALITY of the dingbats showing up is appalling. It doesn't matter if they come from recruiters o by themselves. We are in a period of rapid expansion, the promotions (upward or lateral) "from within" are filled and we can't play musical chairs with the people we already have. And we can't find qualified engineers.
(Clarification: we have about 100 engineers, *not* counting the web guys. I mean real engineers and/or computer scientists and/or equivalent. Also, the really good web guys are equally hard to come by from what I understand.)
Gates' contention in this article is that MS can't find qualified applicants; they aren't after body counts but after quality. We are in a similar position: if qualified people showed up, we'd hire them on-the-spot! We've need at least 30 more engineers and we can't find them. We aren't advocating more H1-Bs, like Gates is. We are, however, desperate to find qualified engineers wherever we can. They can be Indian or Martian for all we care.
H1-B visas are a bandaid for the problem: producing more qualified engineers. We need to create a business and professional environment that promotes the growth of the qualified labor pool HERE. The article, and a lot of posters, are missing this point.
I live in San Francisco. I can't believe that this is happening, but since it is, I have a simple solution: move to another jurisdiction. No, I don't mean "pack your bags and go". I mean that, in this age of interconnected servers throughout the world, hosting your 'blog in another jurisdiction isn't hard to do.
I've ran a couple of servers from a neutral, European country for years. Whenever I want to post something that might piss someone off locally I just post it out of one of those machines and under a pseudonym. While this isn't untraceable by any stretch of the imagination, it makes things hard enough for idiots chasing the poster to give up.
That's the beauty of the Internet/cyberspace. "Here" is simply wherever you want it to be.
I experienced something similar with my Epson Stylus 9000 Color. The printer will report the cartridge as unusuable if you let it there for too long. Epson indicated that the ink degrades over time, yada, yada, yada. I discovered two solutions to this situation:
1. Short term: remove the offending cartridge, wait about 30 seconds, then re-insert the cartridge and run the head cleaning routine. The cartridge will probably work fine.
2. Long term: buy a printer that's on the Laser Monk's list (http://www.lasermonks.com). I've been buying their ink cartridges for a couple of years without problems. I'm about to buy an Epson Stylus R200 -- but I didn't spring for it until I checked that the Monks have the cartridges.
Is it so difficult to understand that OOo is free software, Gnome is free software, KDE is free software and GNU/Linux is free software but OSX is not free software?
So?
Windows isn't free software either and OOo works great in that platform. Mindshare is the name of the game. You can get more mindshare releasing something for Mac than for KDE. I didn't say "don't do the KDE version"; I said "do a Mac version soon". That would get more mindshare than KDE and Gnome combined. For good or evil, everyone always pays attention to what happens in the Apple camps. Whether you like it or not, Apple has been leading the industry for more almost 30 years (yeah, I know about the mid-1990s woes; let's ignore those for now).
I used OOo since the days of StarOffice. I managed to write two books, many presentations, spreadsheets, and countless business documents in it. OOo is probably one of the best office applications and it's cross-platform.
I had quasi undying loyalty to OOo until I decided to go to OS X. While the feature set is almost identical to other versions of OOo, the GUI is one of the ugliest. OOo also lacks compatibility with Exchange servers, which I'm forced to use for work (yuck!). For these two reasons, I had to cave in and return to Office:Mac.
The efforts to tightly couple OOo with KDE or Gnome are important and interesting but far from the marketing win that OOo needs. An OOo version that supports the native OS X look and feel would probably win lots of support from Apple's user base because it would be, in most cases, a drop-in, free replacement for Office:Mac.
I interact now with quite a few Mac users on regular basis; most, if not all, would love to ditch Office:Mac in favour of OOo if the GUI and other system integration issues were resolved. I believe that an OS X/Aqua version of OOo is more strategically important than one for Gnome or KDE because it would generate instant press outside the early-adopter, Linux world.
A strategic marketing win could result in additional funding/participation/donations to OOo to carry on with other projects that, although important, lack the visibility that the Mac has or could bring to OOo.
The DRM techniques discussed in this article sound like an great mental challenge, but nothing else. Every DRM scheme can be defeated if you just stop to think about what you're doing for a few minutes.
Last week we had a slashdot article about the copy protections in currency. I tried scanning a $20 bill with my Canon/Photoshop set up and indeed, Photoshop did not allow the scan to proceed. After tinkering for a bit, it took me the whole of 10 minutes to figure out how to trick my Mac/software/scanner combo to scan the bill. It took me another 5 minutes to feed *the copy* back to the scanner at a resolution that made Photoshop complain that I couldn't scan currency.
(If you're interested, all you need is a 3+ megapixel digital SLR camera, a tripod, and good lighting. Some color matching kung f00 might help. The rest is academic.)
DRM schemes, more than anything, manage to piss people off, and geeks like us just see them as intellectual challenges. DRM hijacking my browser is just the kind of thing that will piss me off and make me want to find a workaround. It would be really sad if this could be circumvented through judicious use of a web proxy, or curl, or wget, or a combination of these, and a Perl script.
Remember also that what is illegal in one country, might be legal elsewhere. I buy all my on-line music through a European broker because they have the same stuff as iTunes and others, plus a lot of European titles hard to get in the US, and offer better prices and no DRM. The same works that Google DRM is trying to block may be easily downloadable, legitimately and without encumbrances, in another jurisdiction. The easiest way to legally hack their DRM might be to simply go through one of their competitors offering the same information without DRM.
Sorry about the rant. I'm just disappointed in Google and everyone else who spend all this effort trying to do DRM and then wind up with egg on their faces when it gets broken by simple methods. Then the issue is handed to the blood sucking lawyers and everyone loses.
The word "android" certainly predates Lucas. Check any dictionary.
The word "droid", on the other hand, was introduced by Lucas/Alan Dean Foster in the first Star Wars movie. Knowing about all the merchandising that resulted from Star Wars, it's not hard to believe that Lucas trademarked it.
Based on my (somewhat limited) knowledge of copyright and trademark law, he's within his right to ask for damages in the case of "droid" if the word is trademarked. Also, I believe you may use the word "android" all day if you want.
How soon before the bad guys set up a dummy corporation and start running nuclear bomb or protein folding simulations on this cluster? I'd be very interested (probably along with some governments) in Weta's and Gen-i screening processes. Will anyone who can foot the bill get access?
I know, this is tinfoil hat stuff, but it's late and I get this "glass half full" visions when I'm sleep deprived.
Yea, because what works for you, works for everyone, right? OS X isn't for everyone. I own a mac and can't stand the thing; I use my thinkpad running debian every day while the mac collects dust.
I also own three Linux/RH enterprise servers, a Sun 420R with Solaris, a notebook running Mandrake 10, a Power Mac G5, and a workstation that runs Windows XP Pro.
The right tool for the right job. I use Macs now almost exclusively for end-user/programming, Linux/Solaris for servers, and XP when someone wants me to do something for Windows. On a Mac, I prefer OS X/Cocoa. On Intel hardware I prefer Linux, usually RH of some sort.
I agree that the PowerBooks are heavier than the mm20 -- even without having seen one of the latter.
Now, about the PCMCIA slots -- PowerBooks don't need them. You have Firewire and almost everything else you might desire is built-in.
As to running Linux on a PowerBook... why? GNU tools are available in there, it runs *NIX, and it has a hell of a nice GUI. Running Linux/KDE on a PowerBook is like hauling hay in a Ferrari. It just doesn't make sense. Linux does a lot of good stuff in PC hardware, but OS X/Darwin works much better under Macs for obvious reasons.
If you dont want one of these, get a Fujitsu P7000 series. I've got a P2046 and the form factor rocks *and* the screen beats out a powerbook's anemic 1024x768 any day.
The Fujitsu P7000 series doesn't have enough memory at the same price, runs Windows XP (yuck) and it's fugly to look at. Chicks won't dig it...
Plus, you get compact flash/sd slots that the powerbook doesn't have.
The PowerBook doesn't need those slots. That's what Firewire is for.
Slashdot Mirror
Greetings.
.Net, depending on problem sub-domain
I work as an architect for one of the largest companies in the world. There have been several attempts at finding the standard language holy grail, and in the end, the strategy that works, is to have a standard language assigned to given problem domains. With that in mind...
* The web site itself is done in Java/JSP/JSTL/etc.
* The business logic is built with PL/SQL
* The scripts that build the site are written in shell, make, m4, Perl
* The QA team gets along marvelously with Python
* The end-user, on-demand digital goods delivery offerings are written
in either Flash or
* Etc.
The policy is to find what's the best tool for a given job, and standardize on that tool. The strategy has worked very well.
My suggestion? The good old "find the right tool for doing a specific job".
Cheers!
Eugene
Sangbin wrote:
"Your list is too long. 1) Breasts Thank you."
He, he, he... if I could sneak that one past HR, of course I would!
Cheers,
E
Greetings,
.Net.
I manage a small but important team. The guys who report to me are, by the definition of their jobs, highly technical. Whenever something complicated needs to be researched and/or implemented, my guys get to do it, especially if it has to do with the adoption of new technologies.
We had our quarterly review a few weeks ago (it goes both ways; they evaluate me, I evaluate them) and the results were excellent. Here are the overall management techniques I employed with them:
1. Hold everyone in the team, including myself, to the highest
standard.
2. Define what 'highest standard' means as a part of the requirements
specification.
3. Once a decision has been made, by the team, business owners, etc.
there is no arguing. Part of my job is to keep the business guys
from becoming a distraction. The other part is to ensure that
the engineers deliver (1) and (2).
4. Go through a quarterly review with them; divide a sheet of paper
in three columns labeled as follows:
a) Desired outcomes (projects, training, coaching of others, etc.)
b) Achievements
c) Areas that need improvement
At the beginning of the quarter first quarter ever that you
implement this, fill-in only items in the first column. At the
end of the quarter, fill in the other two columns. A person is
doing great if they had, say, four desired outcomes and wind up
with four or more achievements. Last, review things that need
improvement (mine is "needs to attend relevant meetings" for this
quarter). Discuss those AND FOCUS ON BEHAVIOUR, not on
personality. Explain why the improvement is needed. After you
negotiate what this means, add it both as a thing to improve and
as a desired outcome for the next quarter. Repeat every quarter.
5. Respect your engineers' decisions. Combined, they know more than
you do, regardless of how technically capable you are. If that's
not the case, you shouldn't be a manager and you're probably not
meeting 1-3.
6. Leave your engineers alone to do what they do best. Don't invite
them to too many meetings or have them do tasks unrelated to their
charter. Engineers hate distractions, and distractions prevent
the team from achieving 1.
7. If the business folks start coming up with eleventh hour changes,
ensure that the engineers are part of the discussion and reason
WITH BOTH SIDES to figure out which changes make sense and why, which
don't, and how to come up with a solution that will meet everyone's
goals. NEVER just inform the engineers that a decision that affects
what they've been working on for three months has been made.
8. As a part of 4, create an environment where you are constantly training
your team, exposing them to new technologies, etc. Reward the intro-
duction of new techniques, procedures, etc. In 4, suggest that they
read at least a new book ON SOMETHING NEW NOT USED AT WORK every quarter.
If you work in a Java shop, they should be reading about Ruby or
You never know when a better mousetrap is available if you aren't informed.
9. Reward excellence whenever you see it, from solving the thorniest algorith-
mic q
As someone who's been programming in Java for almost 10 years and developed several GUI applications and such... yes, I feel like these guys shouldn't have used non-standard stuff that doesn't work out of the box. I don't care if JGoddies is open-source or otherwise. All I care about, as a follow-up poster noted, is that I tried several standard ways of running this in Java and it didn't work. And I blame that JGoodies library because Columba still doesn't work even if JGoodies is in the -cp/CLASSPATH.
.jar files and a prayer to make this run. It didn't work for me and it'll be faster for me to install something else (i.e. Thunderbird) than trying to figure out why this failed.
.sh/.bat file with it that does all the class path magic. A README with instructions on how to run this might help. An installer or a self-contained package would be even better (download and install Thunderbird for a good example of this).
As an end-user, I don't care if it takes ten
Maybe it's just a packaging issue but I find it to be a huge turn off either way. Since it relies on open-source libraries, and they ship them with the code, they could at least ship a
Cheers,
E
Greetings,
1 _0.gif
I just downloaded and tried to configure Columba 1.0 under OS X 10.4.2. My verdict? Skip it.
The people behind Columba used some widget library that's system dependent. This is throwing a number of null pointer exceptions under OS X with the Java 5 JVM. They all relate to something called "jgoodies"; they're doing something that appears to be system dependent.
One of the main reasons for using this would be portability. They seemed to have missed the boat altogether since it doesn't run under an otherwise standad Java configuration! Why bother with writing a Java application if it's not cross-platform? Why use non-standard widget libraries? Attaining cross-portability in Java is hard enough as it is; these guys chose to make it even harder. Thank you for blowing away the only reason I might've had for using the Columba email client.
You can see a screen capture showing the exceptions here:
http://eugeneciurana.com/personal/images/Columba-
Can't say if this works at all because I was unable to tell Columba about my IMAP server. I got another of those jgoodies-related exceptions when I tried to select something other than POP3.
Cheers,
E
Greetings,
A few years ago I learned a trick from my local Swatch store. I had scratched the face of a watch given to me as a gift; a keeper mainly for its sentimental value. The face had several scratches, some looked deep. I steeled myself to pay $50 to replace the watch's face (an $80 watch) and headed to the store.
The clerk was very helpful and passed on one of the best tips ever: Put some toothpaste on the polycarbonate surface, rub softly with your fingers, and wipe off with a moist cotton pad or paper towel. Scratches be gone! I've used that trick on mobile phone screens as well, with excellent results.
I hope that helps,
E
Howdy,
I can't remember the name of the painting off the top of my head, but that's a mink, not a ferret.
The symbolism of the painting is that the woman is so pure that a mink, usually very, very, very fastidious about the cleanliness of its fur, lets her not only touch but even hold it. These little critters don't like *anything* touching them.
If I remember the name of the painting in the morning I'll come back and post a follow-up.
Cheers,
E
Tim,
c ime.net/
You may want to take a look at MontaVista's real-time Linux offerings (http://www.mvista.com./ I was the technical lead for a real-time Java controller core that ran atop Hard Hat Linux and implemented both BACNet and Profibus (via Applicom dedicated I/O cards) for building automation and industrial robots, respectively.
I sold the rights to the software to the oil drilling equipment company that implemented the industrial robots but I'll be happy to assist if you want to discuss. You can find me as pr3d4t0r in irc.freenode.net ##java or chupacabra in Undernet #java.
Here is an old page describing some of the stuff that we did: http://web.archive.org/web/20010302021846/http://
I'll be a speaker at the Java in Action conference in Orlando this coming October; one of the sessions will talk about recent work I made in embedded/mobile/full-automation stuff. Most of my work is now based on Linux with some OS X, Solaris, and QNX to spice things up (I now work full-time for someone else; got tired of the startup game...)
Cheers and good luck,
E
Anonymous Coward wrote:
Seriously, I haven't used a "word processor" in years, except to read stupid infected files other people send me. Spreadsheets are more handy, but Appleworks/OO.o do just fine in this area.
Cheers,
EHowdy.
Interesting comment. For the record, I came here as an H1-B in 1990, before there was too much demand (i.e. I got my green card in less than 18 months). I never had trouble finding work or otherwise keeping busy. The same applies to a few of my friends. And that's my point. It's not about finding the top 1% of anything. It's about finding people who are qualified to tackle a job. And a lot of the people I get to interview I wouldn't hire.
The reality of things is that businesses nowadays have to strive to operate at that top 0.01% or someone else will eat your lunch. It's not pretty, it's not easy, but it's factual. If you don't put 100% into what you're doing, someone else will come and knock you off the perch.
The business that my company is in the most competitive in the world. Unless everyone is pulling in the same direction AND qualified to do their job above mediocrity levels, someone else will come and take our place... and our jobs.
The difference between being a professional and an amateur is in the qualifications. I don't want someone with paper credentials. I want someone who can deliver results for herself and for the company. I don't want the top 0.01%, but I don't want anybody in below the 65% mark either.
(FYI - 65%, no curves, was the passing grade at my university. Anything lower than that was an F. That helped to keep us on our toes.)
I'll think a bit more about your post; I'm a bit too tired to answer more coherently and my answer may not do justice to your comment, which I think it's insightful and will help me in future decisionmaking.
Best wishes,
Eugene
Greetings,
I work as a systems architect for a very large company in Silicon Valley. The problem is not that there are a large number of unemployed programmers out there. There are. The real problem is the lack of qualified programmers if you want to do something serious.
The company I'm working for has about 100 engineers in various capacities. Nobody there has less than five years of experience, nor can we hire anybody with less than that for the work at hand. There are a number of positions open because the QUALITY of the dingbats showing up is appalling. It doesn't matter if they come from recruiters o by themselves. We are in a period of rapid expansion, the promotions (upward or lateral) "from within" are filled and we can't play musical chairs with the people we already have. And we can't find qualified engineers.
(Clarification: we have about 100 engineers, *not* counting the web guys. I mean real engineers and/or computer scientists and/or equivalent. Also, the really good web guys are equally hard to come by from what I understand.)
Gates' contention in this article is that MS can't find qualified applicants; they aren't after body counts but after quality. We are in a similar position: if qualified people showed up, we'd hire them on-the-spot! We've need at least 30 more engineers and we can't find them. We aren't advocating more H1-Bs, like Gates is. We are, however, desperate to find qualified engineers wherever we can. They can be Indian or Martian for all we care.
H1-B visas are a bandaid for the problem: producing more qualified engineers. We need to create a business and professional environment that promotes the growth of the qualified labor pool HERE. The article, and a lot of posters, are missing this point.
Cheers,
Eugene
Greetings.
I live in San Francisco. I can't believe that this is happening, but since it is, I have a simple solution: move to another jurisdiction. No, I don't mean "pack your bags and go". I mean that, in this age of interconnected servers throughout the world, hosting your 'blog in another jurisdiction isn't hard to do.
I've ran a couple of servers from a neutral, European country for years. Whenever I want to post something that might piss someone off locally I just post it out of one of those machines and under a pseudonym. While this isn't untraceable by any stretch of the imagination, it makes things hard enough for idiots chasing the poster to give up.
That's the beauty of the Internet/cyberspace. "Here" is simply wherever you want it to be.
Cheers,
E
D'oh! -- the Monk's is a lot cheaper; that's why I posted that.
Long day, sorry.
Cheers,
Eugene
I experienced something similar with my Epson Stylus 9000 Color. The printer will report the cartridge as unusuable if you let it there for too long. Epson indicated that the ink degrades over time, yada, yada, yada. I discovered two solutions to this situation:
1. Short term: remove the offending cartridge, wait about 30 seconds, then re-insert the cartridge and run the head cleaning routine. The cartridge will probably work fine.
2. Long term: buy a printer that's on the Laser Monk's list (http://www.lasermonks.com). I've been buying their ink cartridges for a couple of years without problems. I'm about to buy an Epson Stylus R200 -- but I didn't spring for it until I checked that the Monks have the cartridges.
I hope this helps.
Cheers,
Eugene
gallir wrote:
Is it so difficult to understand that OOo is free software, Gnome is free software, KDE is free software and GNU/Linux is free software but OSX is not free software?
So?
Windows isn't free software either and OOo works great in that platform. Mindshare is the name of the game. You can get more mindshare releasing something for Mac than for KDE. I didn't say "don't do the KDE version"; I said "do a Mac version soon". That would get more mindshare than KDE and Gnome combined. For good or evil, everyone always pays attention to what happens in the Apple camps. Whether you like it or not, Apple has been leading the industry for more almost 30 years (yeah, I know about the mid-1990s woes; let's ignore those for now).
Cheers,
EExcellent point!
Will that make me an office suite vampire of sorts? The nosferatu of software? Hrm... something to think about...
Thanks for the clarification -- and a good laugh.
Have a great day,
Eugene
Disclaimer: I'm an OOo advocate, as you can see from this Computerworld article (http://www.computerworld.com/softwaretopics/softw are/apps/story/0,10801,92195,00.html?SKC=software- 92195) that I published last Spring.
I used OOo since the days of StarOffice. I managed to write two books, many presentations, spreadsheets, and countless business documents in it. OOo is probably one of the best office applications and it's cross-platform.
I had quasi undying loyalty to OOo until I decided to go to OS X. While the feature set is almost identical to other versions of OOo, the GUI is one of the ugliest. OOo also lacks compatibility with Exchange servers, which I'm forced to use for work (yuck!). For these two reasons, I had to cave in and return to Office:Mac.
The efforts to tightly couple OOo with KDE or Gnome are important and interesting but far from the marketing win that OOo needs. An OOo version that supports the native OS X look and feel would probably win lots of support from Apple's user base because it would be, in most cases, a drop-in, free replacement for Office:Mac.
I interact now with quite a few Mac users on regular basis; most, if not all, would love to ditch Office:Mac in favour of OOo if the GUI and other system integration issues were resolved. I believe that an OS X/Aqua version of OOo is more strategically important than one for Gnome or KDE because it would generate instant press outside the early-adopter, Linux world.
A strategic marketing win could result in additional funding/participation/donations to OOo to carry on with other projects that, although important, lack the visibility that the Mac has or could bring to OOo.
Cheers,
EugeneBig deal. Now, if that had been a free, unencrypted feed of the Spice or Playboy channels...
Cheers!
EGreetings,
The DRM techniques discussed in this article sound like an great mental challenge, but nothing else. Every DRM scheme can be defeated if you just stop to think about what you're doing for a few minutes.
Last week we had a slashdot article about the copy protections in currency. I tried scanning a $20 bill with my Canon/Photoshop set up and indeed, Photoshop did not allow the scan to proceed. After tinkering for a bit, it took me the whole of 10 minutes to figure out how to trick my Mac/software/scanner combo to scan the bill. It took me another 5 minutes to feed *the copy* back to the scanner at a resolution that made Photoshop complain that I couldn't scan currency.
(If you're interested, all you need is a 3+ megapixel digital SLR camera, a tripod, and good lighting. Some color matching kung f00 might help. The rest is academic.)
DRM schemes, more than anything, manage to piss people off, and geeks like us just see them as intellectual challenges. DRM hijacking my browser is just the kind of thing that will piss me off and make me want to find a workaround. It would be really sad if this could be circumvented through judicious use of a web proxy, or curl, or wget, or a combination of these, and a Perl script.
Remember also that what is illegal in one country, might be legal elsewhere. I buy all my on-line music through a European broker because they have the same stuff as iTunes and others, plus a lot of European titles hard to get in the US, and offer better prices and no DRM. The same works that Google DRM is trying to block may be easily downloadable, legitimately and without encumbrances, in another jurisdiction. The easiest way to legally hack their DRM might be to simply go through one of their competitors offering the same information without DRM.
Sorry about the rant. I'm just disappointed in Google and everyone else who spend all this effort trying to do DRM and then wind up with egg on their faces when it gets broken by simple methods. Then the issue is handed to the blood sucking lawyers and everyone loses.
Cheers,
E
The word "android" certainly predates Lucas. Check any dictionary.
The word "droid", on the other hand, was introduced by Lucas/Alan Dean Foster in the first Star Wars movie. Knowing about all the merchandising that resulted from Star Wars, it's not hard to believe that Lucas trademarked it.
Based on my (somewhat limited) knowledge of copyright and trademark law, he's within his right to ask for damages in the case of "droid" if the word is trademarked. Also, I believe you may use the word "android" all day if you want.
Cheers,
E
Hrm...
How soon before the bad guys set up a dummy corporation and start running nuclear bomb or protein folding simulations on this cluster? I'd be very interested (probably along with some governments) in Weta's and Gen-i screening processes. Will anyone who can foot the bill get access?
I know, this is tinfoil hat stuff, but it's late and I get this "glass half full" visions when I'm sleep deprived.
Cheers,
E
defF024 wrote:
Yea, because what works for you, works for everyone, right? OS X isn't for everyone. I own a mac and can't stand the thing; I use my thinkpad running debian every day while the mac collects dust.
I also own three Linux/RH enterprise servers, a Sun 420R with Solaris, a notebook running Mandrake 10, a Power Mac G5, and a workstation that runs Windows XP Pro.
The right tool for the right job. I use Macs now almost exclusively for end-user/programming, Linux/Solaris for servers, and XP when someone wants me to do something for Windows. On a Mac, I prefer OS X/Cocoa. On Intel hardware I prefer Linux, usually RH of some sort.
Cheers
EI agree that the PowerBooks are heavier than the mm20 -- even without having seen one of the latter.
Now, about the PCMCIA slots -- PowerBooks don't need them. You have Firewire and almost everything else you might desire is built-in.
As to running Linux on a PowerBook... why? GNU tools are available in there, it runs *NIX, and it has a hell of a nice GUI. Running Linux/KDE on a PowerBook is like hauling hay in a Ferrari. It just doesn't make sense. Linux does a lot of good stuff in PC hardware, but OS X/Darwin works much better under Macs for obvious reasons.
Cheers,
E
dieman wrote:
If you dont want one of these, get a Fujitsu P7000 series. I've got a P2046 and the form factor rocks *and* the screen beats out a powerbook's anemic 1024x768 any day.
The Fujitsu P7000 series doesn't have enough memory at the same price, runs Windows XP (yuck) and it's fugly to look at. Chicks won't dig it...
Plus, you get compact flash/sd slots that the powerbook doesn't have.
The PowerBook doesn't need those slots. That's what Firewire is for.
Cheers,
E
I forgot: For $1,900 your PowerBook will also have around 768 MB RAM, maybe even 1 GB, vs. the 256 MB offered by the Moebius.
Cheers!
E