IMHO a simple improvement to email would be no more than twice a day delivery.
OMG, are you insane? I get 200 "urgent" emails a day (not with a flag, but issues people actually want a response to soon). If we had twice a day delivery I would have to answer 200 PHONE CALLS a day instead.
On average a Ph.D. (in biology at any rate) will have spent 8 years or so post BS before getting a job at a big Pharmaceutical company. Sure, I know that there are exceptions, but the combination of doctorate and post-doctoral fellowship tends to be around 8 years. Sometimes less, occasionally a great deal more.
Get a B.S. and work hard for 8 years and if you are good you may have the same office and the same pay. Plus you have earned actual money for 8 years while the guy with the Ph.D. (me) hasn't even started saving for retirement yet.
IMO the best education/work balance for big Pharma (again for biology - perhaps also for chemistry) is the M.S. They will have the easiest time getting a job and the couple of years spent getting the M.S. will probably pay for itself.
The downside is that those without a Ph.D. will tend to hit a ceiling earlier in their career. But not always.
My real world experience tells me it is much less than 32%. 15% at best (though another 15% THINK they know java - this is where the real danger lies).
If you require knowledge of complex topics like sensible J2EE architecture or multi-threading it falls into the single digits.
The second half of the article recommends Model Driven Architecture for the masses as the solution. This amounts to putting complex tools into the hands of idiots. Tools that go out of their way to keep people ignorant, while simultaneously giving them the power to commit their sins on a grand scale. Brilliant.
My but there are a lot of negative comments up there (note that I do not necessarily think all of the negative comments are wrong - this really IS a bad time to be entering the industry). But, there are some positive jewels too - go back and re-read No Sweat For You by Cowbert Prime, The dissenting opinion by xenocide2, Combine Your Talents by iplayfast, and Become a craftsman... by Satan's Librarian.
I have a Ph.D. in biology and recently transitioned to a programming job at a major pharmaceutical company. I did this at a time when the market was absolutely flooded with programmers blasting out of the popped bubble. In order to accomplish this I had to be willing to start at the bottom (a low paid contractor) in lieu of demonstrable programming experience, I had to be an excellent programer (and willing to work hard and prove it), I had to combine all of my skills together into a coherent whole, and I had to get a lucky break.
I considered many of the training options that you have and that were suggested in other posts. My plan was to become a Sun Certified Java Programmer as proof that I wasn't a complete technical idiot, do some work on an open source project related to the field, and go from there. I got my lucky break before I took the certification exam, but I believe that it was basically a sound plan.
My science background turned out to be a perfect match for this job. Since I excel in both fields I can tackle problems that no ordinary coder would stand a chance at and no scientist has the time or programming skills for.
As I am now in a position to hire or influence the hiring of people, here are some things I would be looking for if I had your resume on my desk.
A body of programming work (open source projects or just personal projects, but something I could look at and probe you for knowledge on)
Some kind of paper proof that you might know what you are doing (Java Certification or some other comparatively difficult certification, a masters in computer science, etc)
Claimed knowlege of a variety of technologies (Java, XML, HTML, Web Services, J2EE,.NET, etc). The actual mix you would need would depend a great deal on the actual job, but you should show breadth if possible.
If I then interviewed you I would be looking for things like:
Knowledge of industry jargon - particularly jargon that might demonstrate that you are serious about learning everything you can to make yourself a better coder (What do you know about patterns, agile programming, software craftsmanship, etc).
Understanding of all of the things you claim on your resume.
Deep curiosity about technology (a hacker's mind). It is very encouraging that you are reading/.
Excellent problem solving skills combined with a deep need to make it work as well as possible (a coder's mind)
And in your case I would quiz you on basic science as well. How much organic chemistry DO you remember?:-)
I am not trying to sell you on programming for the pharmaceutical industry. This was all just intended to give you a real-world example and to inspire you to find your own niche.
10 Seconds? Sheesh, what version of the plugin are you running?
But no matter. I still claim that in practice (for most people it will not be 10 seconds) this is generally no longer than your average user is going to wait for your average web-app page load unless you have a wicked fast network and it will only take that long the very first time an applet is loaded, so you could even force this loading to occur ahead of time in a multi-page scenario.
Dead-locks in applets are at most as common as multi-threaded applets, which almost don't exist. I am not saying that one can't freeze the VM, I am just saying that it's more likely to be an endless loop of some kind than thread dead-lock per se.
So you do have a point about sharing the JVM, though I have never actally seen my browser VM freeze that I can recall.
If I can't convince you to try applets, then how about web start? If you truly want interactivity, then just add a "click here for interactive graph" link that starts a genuine rich client application. This WOULD require your users to install a modern JVM (the applet route would not), but now there isn't anything you can't do.
If you want the interactivity, you should at least try an optimized pared down applet. It's true that most applets are slow, but it is frequently also true that they are written badly and/or contain heaps of additional libraries that they need to download. Use as much of java's built-in code as you can with the possible exception of the GUI library.
Look at lightweight graphical libraries like lwvcl (commercial) or
thinlet (LGPL) for the controls. The zaval people even have a charting library for their lwvcl system. (I have never used the lwvcl library, it just looks cool - try their online demo. Thinlet packs amazing capabilities into a very small package.)
If you need to display 15 graphs at a time with limited interaction, then this may not be the way to go, but if you need to display one at a time with very rich interaction, this might be the ticket.
Darn it! I wanted to moderate, but I'm going to have to post because everyone seems to have missed something important here. (donates his moderator points to the ether)
The really really scary thing about this is that no one who has been infected is getting better.
That by itself might not be enough to have everyone in an uproar, but throw in moderate mortality and extremely infectious and completely unknown infectious agent and it is definitely time to be concerned.
Just in case you still don't see the implications, imagine a disease that spreads with the same ease as the flu or the common cold, but from which you never recover. OK, that's scary right? I have a cold (or flu) at the moment, so I know that I find this scary.
Now imagine that EVERYONE who has caught it is a carrier of it forever, but they don't tend to die, so there are LOTS of carriers. OMG, very scary.
OK, now imagine that the symptoms can be relatively mild for long periods, meaning people will travel and go to work and to school for quite some time before they seek professional help (and thus be quarantined). Jeez. Too scary for words, really.
How long before a significant fraction of the population of the earth could be infected by such a disease?
Let's just hope the actual disease turns out not to be as bad as my hypothetical one.
Indeed. And whether or not samba gets installed, the programmers can run jedit locally and use the FTP plugin to remotely edit the files.
The FTP plugin is probably the one I use the most. The only disadvantage, really, is that if you are working on something like scripts you will have to chmod the files before you run them.
Slashdot Mirror
As someone going for their PhD, I have to wonder if it actually does.
I absolutely guarantee that none of the people who modded this funny have a Ph.D.
This comment was "insightful" people. Pay attention.
David
IMHO a simple improvement to email would be no more than twice a day delivery.
OMG, are you insane? I get 200 "urgent" emails a day (not with a flag, but issues people actually want a response to soon). If we had twice a day delivery I would have to answer 200 PHONE CALLS a day instead.
I would have to find a cyanide capsule somewhere.
dhk
On average a Ph.D. (in biology at any rate) will have spent 8 years or so post BS before getting a job at a big Pharmaceutical company. Sure, I know that there are exceptions, but the combination of doctorate and post-doctoral fellowship tends to be around 8 years. Sometimes less, occasionally a great deal more.
Get a B.S. and work hard for 8 years and if you are good you may have the same office and the same pay. Plus you have earned actual money for 8 years while the guy with the Ph.D. (me) hasn't even started saving for retirement yet.
IMO the best education/work balance for big Pharma (again for biology - perhaps also for chemistry) is the M.S. They will have the easiest time getting a job and the couple of years spent getting the M.S. will probably pay for itself.
The downside is that those without a Ph.D. will tend to hit a ceiling earlier in their career. But not always.
And anyway, no one calls me Doctor.
dhk
My real world experience tells me it is much less than 32%. 15% at best (though another 15% THINK they know java - this is where the real danger lies).
If you require knowledge of complex topics like sensible J2EE architecture or multi-threading it falls into the single digits.
The second half of the article recommends Model Driven Architecture for the masses as the solution. This amounts to putting complex tools into the hands of idiots. Tools that go out of their way to keep people ignorant, while simultaneously giving them the power to commit their sins on a grand scale. Brilliant.
dhk
I have a Ph.D. in biology and recently transitioned to a programming job at a major pharmaceutical company. I did this at a time when the market was absolutely flooded with programmers blasting out of the popped bubble. In order to accomplish this I had to be willing to start at the bottom (a low paid contractor) in lieu of demonstrable programming experience, I had to be an excellent programer (and willing to work hard and prove it), I had to combine all of my skills together into a coherent whole, and I had to get a lucky break.
I considered many of the training options that you have and that were suggested in other posts. My plan was to become a Sun Certified Java Programmer as proof that I wasn't a complete technical idiot, do some work on an open source project related to the field, and go from there. I got my lucky break before I took the certification exam, but I believe that it was basically a sound plan.
My science background turned out to be a perfect match for this job. Since I excel in both fields I can tackle problems that no ordinary coder would stand a chance at and no scientist has the time or programming skills for.
As I am now in a position to hire or influence the hiring of people, here are some things I would be looking for if I had your resume on my desk.
- A body of programming work (open source projects or just personal projects, but something I could look at and probe you for knowledge on)
- Some kind of paper proof that you might know what you are doing (Java Certification or some other comparatively difficult certification, a masters in computer science, etc)
- Claimed knowlege of a variety of technologies (Java, XML, HTML, Web Services, J2EE,
.NET, etc). The actual mix you would need would depend a great deal on the actual job, but you should show breadth if possible.
If I then interviewed you I would be looking for things like:I am not trying to sell you on programming for the pharmaceutical industry. This was all just intended to give you a real-world example and to inspire you to find your own niche.
dhk
The jikes that you "have unknowingly acually used" was most likely just the compiler (which has been available for ages and is excellent).
It probably was not the virtual machine described here.
10 Seconds? Sheesh, what version of the plugin are you running?
But no matter. I still claim that in practice (for most people it will not be 10 seconds) this is generally no longer than your average user is going to wait for your average web-app page load unless you have a wicked fast network and it will only take that long the very first time an applet is loaded, so you could even force this loading to occur ahead of time in a multi-page scenario.
Dead-locks in applets are at most as common as multi-threaded applets, which almost don't exist. I am not saying that one can't freeze the VM, I am just saying that it's more likely to be an endless loop of some kind than thread dead-lock per se.
So you do have a point about sharing the JVM, though I have never actally seen my browser VM freeze that I can recall.
If I can't convince you to try applets, then how about web start? If you truly want interactivity, then just add a "click here for interactive graph" link that starts a genuine rich client application. This WOULD require your users to install a modern JVM (the applet route would not), but now there isn't anything you can't do.
dhk
If you want the interactivity, you should at least try an optimized pared down applet. It's true that most applets are slow, but it is frequently also true that they are written badly and/or contain heaps of additional libraries that they need to download. Use as much of java's built-in code as you can with the possible exception of the GUI library.
Look at lightweight graphical libraries like lwvcl (commercial) or thinlet (LGPL) for the controls. The zaval people even have a charting library for their lwvcl system. (I have never used the lwvcl library, it just looks cool - try their online demo. Thinlet packs amazing capabilities into a very small package.)
If you need to display 15 graphs at a time with limited interaction, then this may not be the way to go, but if you need to display one at a time with very rich interaction, this might be the ticket.
dhk
Darn it! I wanted to moderate, but I'm going to have to post because everyone seems to have missed something important here. (donates his moderator points to the ether)
The really really scary thing about this is that no one who has been infected is getting better.
That by itself might not be enough to have everyone in an uproar, but throw in moderate mortality and extremely infectious and completely unknown infectious agent and it is definitely time to be concerned.
Just in case you still don't see the implications, imagine a disease that spreads with the same ease as the flu or the common cold, but from which you never recover. OK, that's scary right? I have a cold (or flu) at the moment, so I know that I find this scary.
Now imagine that EVERYONE who has caught it is a carrier of it forever, but they don't tend to die, so there are LOTS of carriers. OMG, very scary.
OK, now imagine that the symptoms can be relatively mild for long periods, meaning people will travel and go to work and to school for quite some time before they seek professional help (and thus be quarantined). Jeez. Too scary for words, really.
How long before a significant fraction of the population of the earth could be infected by such a disease?
Let's just hope the actual disease turns out not to be as bad as my hypothetical one.
-dhk-
Indeed. And whether or not samba gets installed, the programmers can run jedit locally and use the FTP plugin to remotely edit the files. The FTP plugin is probably the one I use the most. The only disadvantage, really, is that if you are working on something like scripts you will have to chmod the files before you run them.