The TLDR version is this scenario is why you configure your mythtv box to store MPEG TS which have embedded CRC error detection and recovery instead of MPEG PS which are irrelevantly smaller, if you have the option.
That won't help detect corruption, only truncation of files. You would need an md5 or similar hash.
md5 is (relatively) slow. a simple CRC-32 will only fail you for 1 in 2 ** 32 corruptions, and I suspect the guy doesn't even have 2 ** 16 files so the odds are CRC-32 is more than good enough and significantly faster.
Then again, he's probably going to be hard drive speed limited not CPU limited. Then again, no point wasting laptop battery on an overly complicated algorithm. CRC32 is gonna use at least 1/5th the CPU/wallclock time and/or battery of md5.
The tradeoff boils down to you can use md5 and burn at least 5 times more battery/heat/wall clock time (whatever is your limiting reagent) in exchange for (128-32) = 2 ** 96 times lower likelihood of mistake. The problem with accepting 2 ** 96 higher reliability is his dying hard drive probably cannot provide 2 ** 16 reliability so increasing the algorithm is a waste since it's already asymptotically limited.
unix "file" is not the answer. For some formats it does as little as look at a couple header bytes. Its a great tool to guess a format. Its a terrible verifying parser and does nothing to verify content.
An example of what I'm getting at, with some made up details, unfortunately html is not like well formed xml and every viewer is different anyway so the best way to figure out if a html web page file format is corrupt is unfortunately to pull it up in firefox. This only detects corruption in the structure of the file, if the corruption is just a couple bits then you end up with problems like tQis where the only way to see the h got fouled up is to write more or less a IQ 100 artificial intelligence. All "file" is going to test is pretty much does the file begin with or contain a regex something like less-than html greater-than (getting past the filters).
For content you could F around with, for example, piping a mp3 file thru a decoder and then thru an averaging spectrum analyzer and see if there's anything overly unusual in the spectrum. Also some heuristics like is the file only 1 second long, then its F'ed up.
I don't have one, but I thought the kinect did 2D very accurately plus a crude 3rd D based on image size so lets call it 2.5 D
I don't see how one mic and two speakers does more than 1 D of data. Then again I haven't read the article, maybe they place the whole laptop on an oscillating fan or something as a gimmick. Or is it really using the built in cam and the ultrasound is the gimmick that doesn't really do anything?
You've just described my google+ experience so far.
Ahh but see
I'm REALLY unimpressed. The most important and influential web 2.0 company is used by... practically no one. The emperor has no clothes!
The situation is similar, yet the G+ gets endless trash talking about how irrelevant it is and twitter gets endless trash talking about how important and influential it is. That is the difference.
I've given up on figuring out a world wide IT/programmer counts. Best I could figure is github is well over a million users (not projects, but registered user count), we'll say that twitter is at most only 20 times more influential than github... however... there are a lot more programmers than github users.
I think I have enough circumstantial data that I can comfortably stand by my claim that there are more # using programmers than # using twits.
Twitter is a world wide service for mobile phone owners.
World population is about 7 billion.
According to wikipedia: "In February 2010, there were 5.6 billion mobile phone subscribers, a number that is expected to grow." This seems bogus high... there are people in 1st world with both business and personal phones, of course, but that would imply there are people who have no food, no water, no shelter, no medical care, yet pay a monthly phone subscription fee. hmm.
So somewhere around 70% of the worlds population is theoretically a possible twitter user. Lets see how much of that 70% use twitter.
Most popular twitter account is lady gaga with 23 million.
So somewhere around a third of one percent of the worlds population follows lady gaga on twitter. I'm REALLY unimpressed. The most important and influential web 2.0 company is used by... practically no one. The emperor has no clothes!
I search google for "IT worker count worldwide" and the third hit to come up is wikipedia's "Discouraged worker" article, which says a lot and is a whole nother story.
Anyway what I'm trying to figure out is how 23 million compares to the number of worldwide programmers/sysadmins/unix users. I theorize that "we" as/. readers own the "#" not twitter. My theory is there are more people that type #!/bin/bash on a daily basis that send tweets on a daily basis. Daily basis is important... I technically have a twitter account, I sent about 2 tweets, subscribed to some morons and some PR agents fronting for some media people, watched for awhile, said WTF is this and never used it again. I would assume this is a rather large fraction of their "subscribers".
This has certain artistic implications for a # skyscraper. Like it needs a neckbeard, or suspenders, or a penguin themed radio antenna on top holding a wifi pringles can... Its not a twitter building because basically no one uses twitter, they're a rounding error.
I will give them credit that there are more twitter users than Intercal programmers using the # operator. Of course if you calculate the sum of the IQs of the two groups the ratio probably narrows a bit..
Donno. The HP killer was the impedance mismatch between ultra high volume commodity computer sales for a profit of $10K/shipping container, medium volume selling small quantities of inkjet ink for $10K/gallon, and ultra low volume selling exotic EE test equipment for roughly $10K/kilo. If you found three independent companies and pitched for VC or junk bonds to finance a merger, they'd put you in a straitjacket. No matter who was in charge, HP was going to have to explode and crash, and the people who thought they were black forest gnomes crafting the worlds best solid platinum coo coo clocks were going to be pissed off if they ended up in the imploding sweatshop shovelware division. Swap em and I think HP is just as dead with Jobs in charge. He might have been that 0.1% miracle worker class, but it was not a survivable situation. It did not help that Fiorina could not lead starving dogs to raw meat and was apparently hired solely as an affirmative action / political move.
There's a slight difference between IT in 1979 and IT in 2012.
Like what? I've only been in the biz for $$$ since 1992 although I've been doing home computing since 81 and its always been in the family. I can't think of anything profoundly new.
I would bet all you can give is fuzzy stuff like "certain numbers are larger" (who cares just say TB instead of MB and you'll be having the same conversations) and perhaps some waffly stuff about mobile platforms which don't really change any of the problems or problem solving techniques, just changes the players.
IT is a game of working around and within limitations. I will give you that the exact ratios of storage space and processing speed and especially latency have varied, but not enough to fundamentally change the game or create anything new.
Gladiatorial combat is still gladiatorial combat, even if you have a new generation of fighters and the arena has been recently remodeled, and even if the swords are an inch longer or shorter than a generation ago.
Please don't tell me you think vmware recently invented the concept of virtualization, or Sun/Java recently invented tokenized/pcode and/or chroot jails. Oh wait, the service bureau known as cloud computing... We had widespread RAID, we just called is DASD... We had cluster computing under VMS and IBM products.
Most importantly, we had the same fundamental problems and tradeoffs as today. The manufacturers and some specification numbers have changed. Nothing has fundamentally changed in the biz since then.
Did you verify this? Did you check with any of their other clients? Did you get examples of their work?
Did you set up delivery schedules? Did you send anyone to India to "sniff around"?
Le me guess... "Why should we? The outsourcing consultant who made a lot of money said it would be just like hiring local employees and we've never done any of that to our employees."
If the burning point of the epoxy is higher than the melting point of the substrate, you can just print over it and melt on another layer, I suppose. Epoxy is a thermoset, it doesn't melt, just burns above 200 degrees C or so (depends on type).
GOOG drive: great online office apps, can also store other stuff. No linux client (so far, or maybe forever, who knows). More space that DB
Dropbox: No built in file viewer/editor for office type apps. Excellent free linux client (no kidding, just install, start, it works. No memory leaks. No bugs. No weirdness. Just freaking works. Nice job guys)
Analysis: dropbox is the base product, beaten across all fronts. GOOG Drive is online office storing on the drive, now storing any file you'd like too. GOOG also has more space. If you use linux you can't use goog drive, so I don't, otherwise it beats dropbox across the board.
Both have nuts TOS etc, so just act like you're posting everything public to the whole world, and assume they'll steal ownership of anything you give them access to. Oddly enough, this doesn't reduce their usefulness very much at all, at least to me.
Online storage is a commodity. Its purefanboyism or audio-phoolism to claim your 1s and 0s sound better if you store your bits on Seagate drives or Western Digital drives. You're better off claiming that drawing a green marker on your ethernet cable makes your cloud stored mp3s sound better. Ditto GOOG or DB for storing 1s 0s over IP. Sounds like GOOG wins across the board unless you've got linux then its unusable so DB as the only serious entrant wins.
looking more like a good manuscript 700 years ago than a magazine piece only 10 years ago
What does this even mean? I can't parse it and its early Sunday morning but I'm not that drowsy or stupid (although some would disagree with the latter).
Obviously its relevant that he references GOOD 700 year old vs just "a" 10 year old. Obviously its important that magazines are time filling fluff for the masses / chewing gum for the mind, and in the old days manuscripts held real individual contributions of science work (like a modern journal / preprint archive / e-journal). Manuscripts didn't do much for complimentary copy other than the occasional thank god here and there, but magazines are almost all complimentary copy now?
So instead of using that icky earth destroying copper wire, it uses conductive thread. Thread is used by girls making craft projects, what an excellent correlation with the marketing campaign that only women like health and green (seriously, WTF is up with that?). Tada, conductive thread, its great!
Seriously, conductive thread is basically wire wrapping wire with yarn/thread except the connections aren't gas tight so its not as reliable. Wirewrap is great stuff, I built a 8051 based microcontroller in '91 and it still works. Its a 8052AH-BASIC which is basically a preprogrammed 8051, predating the identical concept BASIC-STAMP and more modern ARDUINO by a decade or two. Wire wrap is the opposite of automotive/industrial/aerospace grade as it is completely intolerant of vibration and moisture. Aside from that, its great. I would guess conductive thread would be the same.
Reading the articles, its a cool psuedo Manhattan style construction using little pegs and making the electrical contacts using the psuedo wirewrap thread. I like manhattan construction for experimental stuff... need another connection point? snap off a tiny piece of DS PCB and solder it to the groundplane...
Simplest similar design would be a 3-d printer that can print Kapton filament and regular ole solder paste and a hot air gun. One problem being that a lot of repraps use Kapton for their high temp parts, so you'd need something more exotic. Aside from my having no idea if Kapton is thermo-setting or thermo-plastic and being too lazy to look it up because it ain't happening anytime soon.
If you don't want to use conductive thread, silver bearing epoxy would probably work.
I have been involved in projects where part of the team was outsourced in India and you often get what you paid for with outsourcing from what I have seen.
Let me guess
"I can hire a whole team of educated experienced developers in India for the cost of one American dev"
turns into a bunch of imaginary billing by people with fake degrees and no experience and no communication so it ends up costing more and not working.
Analogy: Tenth cost android tablet in India ends up late and doesn't work. Isn't this Just how it always turns out over there?
This is a nice idea, but it flies directly into the extremely strongly held cultural / sociological belief that there is no difference between education and training and they're just synonyms for the same thing. You'd have better luck convincing people God does not exist thru logical argument. "College is training for a good job" is as closely held a belief as "god exists" In some ways, more closely held.
We have badges, they're called certifications, and decades of handing them out like crackerjack prizes means they're mostly useless. Certs with an industry wide accreditation board would work, and to the best of my knowledge has never been tried because of the excuse that developing classes, books, and tests means the cert designers would need to be involved in early blue sky development (as if that were not a pile of B.S. and as if NDAs didn't exist and as if the accreditation board needs to be involved at the earliest level of certification)
Back in 1998 a CCNA meant instant job offer. In 2012 I heard some guys at work talking about getting one because they're a ticket to success. It becomes a belief system, like people who still think getting a law degree means instant millions despite 50% unemployment for law grads for years and years now. If you ever get a badge/cert program in, people will believe in it for decades after it closes down, so maybe a high barrier to entry makes sense after all.
The funny thing is nothing is ever new in IT... Its all the same old stuff with new marketing, as the natural cycle turns. There is a huge competitive on the job advantage in having experienced the previous cycle(s) which noob IT people are completely blind to.
Also its impossible to be a good software engineer or good sysadmin without knowing CS. Almost an oxymoron. Maybe they don't know enough math to follow Knuth, but a good one has at least a gut level instinctual level of low level CS knowledge. Its always hilarious for me to watch IT noobs learn the CS guys not only have formulae and charts and provable theorems about their little scalability problem, but their 2010 gut level guesses and hard fought failed experiments were all figured out by the CS guys in the 60s/70s if they'd only have bothered to learn.
While the line isn't always clear, in general it's NOT OK to lie on resume to obtain a job or gain advancement
I cry bogus. When the economy results in 100 fully qualified applicants for each job, the only way to rise to the top of the resume pile is to lie. Therefore most hired by resume and resume filtration are liars, or at least the percentage of liars is spectacularly high, or honest people are dramatically underemployed. I've gotten all my jobs since 1995 thru "knowing people" and "having heard about me" so I haven't had to lie, I've got no dog in the fight so I can be honest about the situation.
A good resume is a legalistically written document full of unprovable lies.
Seriously, doesn't every resume or cover letter contain BS about being "a team player" and being the cause of incredible business achievements? If all of that crap were true, we'd all be surrounded by superheros, or at least superheros who are mysteriously all retired-in-place. Since we aren't, that would inductively seem to indicate most resumes are works of fiction, at best historically accurate or historically inspired fiction.
Example: I've programmed in Perl for $$ since 1996. I know enough about Perl and CPAN to know its a bigger ecosystem than any human being can really understand in totality even if that is their full time academic job. To me, writing "I know Perl" on my resume feels OK, but I know what it takes to be an expert and I am not an expert because I spend time doing other stuff, other languages, etc. There are a lot of clowns out there who once glanced at a Oreilly book on a shelf and maybe changed a couple strings in a downloaded script who write "I am an expert at Perl" on their resume. One of us is lying, but its a somewhat subjective judgment. To a HR chick, we both know a heck of a lot more Perl than she does, so its technically true from her perspective, and disproving it is going to take an interview by someone with specialized knowledge. The liar sounds like a good interview to schedule to a HR chick. I'll never get that interview from HR because the "expert" resume will rise to the top of the resume pile (well, also ageism means I'm disqualified from any future IT employment by HR, unless I lie on my resume by forward dating graduation years, skipping over a couple jobs, etc... thankfully I don't have gray hair and no wrinkles because I don't suntan). Possibly, if I know the hiring manager and maybe a couple of his underlings, I can bypass HR, get an interview, and get the job despite being "too old" and "not being an expert". That's pretty much how its worked so far.
What is wrong is getting caught by stuffing the resume with stuff that's easily dis-proven. Go ahead, prove I wasn't the NCOIC of my computer control section when I was still in the Army in June of 1995, which is kind of a staffing anomaly for a lowly E-4 (usually a E-6 SSG slot). That story is true, but how the hell you'd prove I was lying is a complete mystery to me. I think they put me there as a test, so both I and them could see if I was going to be a lifer. Answer, a mutually agreed upon no. It was fun, and nothing bad happened, but, no longer interested, thanks. Job assignment should be more like that.
The only constant of corporate life is groupthink concentrates at the top. A standard issue stuffed suit is identical to any other standard issue stuffed suit. Any variation in results by different stuffed suits is caused by natural market variation. Who the board selects is not really all that important. Very much like programmer output, the top 0.1% of rock star workers (programmer or CEO) will outperform the masses by a factor of 10, but there are not enough 0.1% rock stars to really matter.
On the other hand, if you have a peon 1st level cust service who lies all the time, they can destroy millions in shareholder value per week, if they try hard enough.
Take a look a the third or so pic on http://lab.dereklow.co/brad/ The one with the light switch hanging out of the wall as he screws around with the wiring. If he could have just stuck to plug in modules like everyone else, but no he has to go all amateur electrician... I love this quote "With no access to the circuit breakers of the dorms, the contacts and wiring inside the wall switch remains live even as I open and try to modify it.". This dude is the stereotype of nothing is more dangerous than a programmer with a screwdriver... damn...
Screwing around with the drapes might have pissed them off too. Worst case is losing some security deposit unless he can return it all exactly to original operation.
Oh and the fog machine. My dorm didn't allow gasoline and oil products in the rooms, to discourage people from putting their scooter in the room, or doing oil changes in the dorms. Maybe they're worked up about fog juice and treating it as automotive lubrication oil.
If you're not experiencing the joys of doin' it the hard way, and you don't care if it can't be scaled above 200 million users per site, why not just string together a bunch of CPAN modules and be done in a weekend? That's the part I couldn't figure out.
RoR forces you into certain mindsets for doing things that may make life harder, and the ruby library infrastructure, while good, isn't as wide ranging or as well debugged as CPAN.
I kinda like all 3 languages mentioned, it just seems Perl would have been the correct tool for this particular job.
The funny part is RoR has recently (well, recent YEARS) gone thru the growing pains C++ suffered thru maybe a decade or so ago. C++ used to be exactly the same way, you upgrade gcc/g++ and suddenly nothing compiles anymore. I think you're pretty much past that with RoR so its safe?
I think they were also trying to make a political point about scalability. The point of the diaspora is not to centralize, not to scale. There were people who Just Didn't Get It about diasporas goals complaining about RoR "How you gonna scale one RoR site to over 500 million users" etc. Thats, um, kind of the point. Shouldn't have more than a household or so per server. Not enough people per server that the admin could make serious dough selling the private info of people he doesn't care about but are on his server anyway. A self correcting privacy policy, sorta, via dispersal.
Hmm my point was not just fire the bottom half, but to realize that what a smart/.er would consider blindingly obvious might be far beyond the ability of perhaps the bottom half of managers, especially the lower level ones.
Sometimes it sucks to work for a guy who got his job solely based on his golf game, or who his daddy is... that's your fault for not gaming it into a new scenario where it rocks to work for someone who got his job solely based on his golf game, or who his daddy is... In the long run that behavior will destroy the company, but no one cares about the long run anymore, so...
Ah thats boring. Wait until it gets exciting, like plumbing leaks, or roof needs replacement. For extra fun, imagine court case involving the rental property. Or insurance claim (storm damage, etc). Another good entertainment is local code enforcement. Finally I'll assume that as a/.er you're pretty respectable, but some renters... are not. you can really work a guy across the country...
Now there are property management companies that will take care of these kind of problems... for a fee, often rather high. Your $1600 check is nice, but until you factor in the costs of ownership, prop tax, maintenance can't be deferred forever especially if its rental property, repairs, mortgage if any...
Easier for it to just show up in the mail. I'm very, very, lazy.
Then you're going to be pissed off that several of my reports are now email delivered. No snail mail, or less, anyway.
Examples:
Wisconsin energy is a paper only company. I believe email delivery is optional for quarterlies, but they like to snail mail a big ole paperback book once a year for annual.
I believe PRPFX (a mutual fund) is a email only company. I might have gotten an annual in the snail mail, but I haven't seen a quarterly from those guys in.. forever I guess.
Definitely SLV and GLD (ETFs) are email only companies. I haven't seen paper from those guys since silver was $7 and gold was $500 (which incidentally is pretty close to my DCA share price)
I believe this is a situation where they're legally required to snail mail upon request to a stockholder, so if you really wanted you could send them snail mail and they'd reply with a nice book, in case your wood stove needed kindling. You should make the shareholder services office earn their pay...
I would guess FB will just publish the data to a page and tell people to friend it.
Slashdot Mirror
The TLDR version is this scenario is why you configure your mythtv box to store MPEG TS which have embedded CRC error detection and recovery instead of MPEG PS which are irrelevantly smaller, if you have the option.
That won't help detect corruption, only truncation of files. You would need an md5 or similar hash.
md5 is (relatively) slow. a simple CRC-32 will only fail you for 1 in 2 ** 32 corruptions, and I suspect the guy doesn't even have 2 ** 16 files so the odds are CRC-32 is more than good enough and significantly faster.
Then again, he's probably going to be hard drive speed limited not CPU limited. Then again, no point wasting laptop battery on an overly complicated algorithm. CRC32 is gonna use at least 1/5th the CPU/wallclock time and/or battery of md5.
The tradeoff boils down to you can use md5 and burn at least 5 times more battery/heat/wall clock time (whatever is your limiting reagent) in exchange for (128-32) = 2 ** 96 times lower likelihood of mistake. The problem with accepting 2 ** 96 higher reliability is his dying hard drive probably cannot provide 2 ** 16 reliability so increasing the algorithm is a waste since it's already asymptotically limited.
unix "file" is not the answer. For some formats it does as little as look at a couple header bytes. Its a great tool to guess a format. Its a terrible verifying parser and does nothing to verify content.
An example of what I'm getting at, with some made up details, unfortunately html is not like well formed xml and every viewer is different anyway so the best way to figure out if a html web page file format is corrupt is unfortunately to pull it up in firefox. This only detects corruption in the structure of the file, if the corruption is just a couple bits then you end up with problems like tQis where the only way to see the h got fouled up is to write more or less a IQ 100 artificial intelligence. All "file" is going to test is pretty much does the file begin with or contain a regex something like less-than html greater-than (getting past the filters).
For content you could F around with, for example, piping a mp3 file thru a decoder and then thru an averaging spectrum analyzer and see if there's anything overly unusual in the spectrum. Also some heuristics like is the file only 1 second long, then its F'ed up.
I don't have one, but I thought the kinect did 2D very accurately plus a crude 3rd D based on image size so lets call it 2.5 D
I don't see how one mic and two speakers does more than 1 D of data. Then again I haven't read the article, maybe they place the whole laptop on an oscillating fan or something as a gimmick. Or is it really using the built in cam and the ultrasound is the gimmick that doesn't really do anything?
You've just described my google+ experience so far.
Ahh but see
I'm REALLY unimpressed. The most important and influential web 2.0 company is used by ... practically no one. The emperor has no clothes!
The situation is similar, yet the G+ gets endless trash talking about how irrelevant it is and twitter gets endless trash talking about how important and influential it is. That is the difference.
I've given up on figuring out a world wide IT/programmer counts. Best I could figure is github is well over a million users (not projects, but registered user count), we'll say that twitter is at most only 20 times more influential than github... however... there are a lot more programmers than github users.
I think I have enough circumstantial data that I can comfortably stand by my claim that there are more # using programmers than # using twits.
Who owns #? My theory is twitter does not.
Twitter is a world wide service for mobile phone owners.
World population is about 7 billion.
According to wikipedia: "In February 2010, there were 5.6 billion mobile phone subscribers, a number that is expected to grow." This seems bogus high... there are people in 1st world with both business and personal phones, of course, but that would imply there are people who have no food, no water, no shelter, no medical care, yet pay a monthly phone subscription fee. hmm.
So somewhere around 70% of the worlds population is theoretically a possible twitter user. Lets see how much of that 70% use twitter.
Most popular twitter account is lady gaga with 23 million.
So somewhere around a third of one percent of the worlds population follows lady gaga on twitter. ... practically no one. The emperor has no clothes!
I'm REALLY unimpressed. The most important and influential web 2.0 company is used by
I search google for "IT worker count worldwide" and the third hit to come up is wikipedia's "Discouraged worker" article, which says a lot and is a whole nother story.
Anyway what I'm trying to figure out is how 23 million compares to the number of worldwide programmers/sysadmins/unix users. I theorize that "we" as /. readers own the "#" not twitter. My theory is there are more people that type #!/bin/bash on a daily basis that send tweets on a daily basis. Daily basis is important... I technically have a twitter account, I sent about 2 tweets, subscribed to some morons and some PR agents fronting for some media people, watched for awhile, said WTF is this and never used it again. I would assume this is a rather large fraction of their "subscribers".
This has certain artistic implications for a # skyscraper. Like it needs a neckbeard, or suspenders, or a penguin themed radio antenna on top holding a wifi pringles can... Its not a twitter building because basically no one uses twitter, they're a rounding error.
I will give them credit that there are more twitter users than Intercal programmers using the # operator. Of course if you calculate the sum of the IQs of the two groups the ratio probably narrows a bit..
Donno. The HP killer was the impedance mismatch between ultra high volume commodity computer sales for a profit of $10K/shipping container, medium volume selling small quantities of inkjet ink for $10K/gallon, and ultra low volume selling exotic EE test equipment for roughly $10K/kilo. If you found three independent companies and pitched for VC or junk bonds to finance a merger, they'd put you in a straitjacket. No matter who was in charge, HP was going to have to explode and crash, and the people who thought they were black forest gnomes crafting the worlds best solid platinum coo coo clocks were going to be pissed off if they ended up in the imploding sweatshop shovelware division. Swap em and I think HP is just as dead with Jobs in charge. He might have been that 0.1% miracle worker class, but it was not a survivable situation. It did not help that Fiorina could not lead starving dogs to raw meat and was apparently hired solely as an affirmative action / political move.
There's a slight difference between IT in 1979 and IT in 2012.
Like what? I've only been in the biz for $$$ since 1992 although I've been doing home computing since 81 and its always been in the family. I can't think of anything profoundly new.
I would bet all you can give is fuzzy stuff like "certain numbers are larger" (who cares just say TB instead of MB and you'll be having the same conversations) and perhaps some waffly stuff about mobile platforms which don't really change any of the problems or problem solving techniques, just changes the players.
IT is a game of working around and within limitations. I will give you that the exact ratios of storage space and processing speed and especially latency have varied, but not enough to fundamentally change the game or create anything new.
Gladiatorial combat is still gladiatorial combat, even if you have a new generation of fighters and the arena has been recently remodeled, and even if the swords are an inch longer or shorter than a generation ago.
Please don't tell me you think vmware recently invented the concept of virtualization, or Sun/Java recently invented tokenized/pcode and/or chroot jails. Oh wait, the service bureau known as cloud computing... We had widespread RAID, we just called is DASD... We had cluster computing under VMS and IBM products.
Most importantly, we had the same fundamental problems and tradeoffs as today. The manufacturers and some specification numbers have changed. Nothing has fundamentally changed in the biz since then.
Did you verify this? Did you check with any of their other clients? Did you get examples of their work?
Did you set up delivery schedules? Did you send anyone to India to "sniff around"?
Le me guess... "Why should we? The outsourcing consultant who made a lot of money said it would be just like hiring local employees and we've never done any of that to our employees."
If the burning point of the epoxy is higher than the melting point of the substrate, you can just print over it and melt on another layer, I suppose.
Epoxy is a thermoset, it doesn't melt, just burns above 200 degrees C or so (depends on type).
So how is this different than the modern magazine 10 years ago? You take a newswire story, add some local color, someones blog mentions it, etc.
Doesn't seem to know much about the products.
GOOG drive: great online office apps, can also store other stuff. No linux client (so far, or maybe forever, who knows). More space that DB
Dropbox: No built in file viewer/editor for office type apps. Excellent free linux client (no kidding, just install, start, it works. No memory leaks. No bugs. No weirdness. Just freaking works. Nice job guys)
Analysis: dropbox is the base product, beaten across all fronts. GOOG Drive is online office storing on the drive, now storing any file you'd like too. GOOG also has more space. If you use linux you can't use goog drive, so I don't, otherwise it beats dropbox across the board.
Both have nuts TOS etc, so just act like you're posting everything public to the whole world, and assume they'll steal ownership of anything you give them access to. Oddly enough, this doesn't reduce their usefulness very much at all, at least to me.
Online storage is a commodity. Its purefanboyism or audio-phoolism to claim your 1s and 0s sound better if you store your bits on Seagate drives or Western Digital drives. You're better off claiming that drawing a green marker on your ethernet cable makes your cloud stored mp3s sound better. Ditto GOOG or DB for storing 1s 0s over IP. Sounds like GOOG wins across the board unless you've got linux then its unusable so DB as the only serious entrant wins.
looking more like a good manuscript 700 years ago than a magazine piece only 10 years ago
What does this even mean? I can't parse it and its early Sunday morning but I'm not that drowsy or stupid (although some would disagree with the latter).
Obviously its relevant that he references GOOD 700 year old vs just "a" 10 year old. Obviously its important that magazines are time filling fluff for the masses / chewing gum for the mind, and in the old days manuscripts held real individual contributions of science work (like a modern journal / preprint archive / e-journal). Manuscripts didn't do much for complimentary copy other than the occasional thank god here and there, but magazines are almost all complimentary copy now?
So instead of using that icky earth destroying copper wire, it uses conductive thread. Thread is used by girls making craft projects, what an excellent correlation with the marketing campaign that only women like health and green (seriously, WTF is up with that?). Tada, conductive thread, its great!
Seriously, conductive thread is basically wire wrapping wire with yarn/thread except the connections aren't gas tight so its not as reliable. Wirewrap is great stuff, I built a 8051 based microcontroller in '91 and it still works. Its a 8052AH-BASIC which is basically a preprogrammed 8051, predating the identical concept BASIC-STAMP and more modern ARDUINO by a decade or two. Wire wrap is the opposite of automotive/industrial/aerospace grade as it is completely intolerant of vibration and moisture. Aside from that, its great. I would guess conductive thread would be the same.
Reading the articles, its a cool psuedo Manhattan style construction using little pegs and making the electrical contacts using the psuedo wirewrap thread. I like manhattan construction for experimental stuff... need another connection point? snap off a tiny piece of DS PCB and solder it to the groundplane...
Simplest similar design would be a 3-d printer that can print Kapton filament and regular ole solder paste and a hot air gun. One problem being that a lot of repraps use Kapton for their high temp parts, so you'd need something more exotic. Aside from my having no idea if Kapton is thermo-setting or thermo-plastic and being too lazy to look it up because it ain't happening anytime soon.
If you don't want to use conductive thread, silver bearing epoxy would probably work.
I have been involved in projects where part of the team was outsourced in India and you often get what you paid for with outsourcing from what I have seen.
Let me guess
"I can hire a whole team of educated experienced developers in India for the cost of one American dev"
turns into a bunch of imaginary billing by people with fake degrees and no experience and no communication so it ends up costing more and not working.
Analogy: Tenth cost android tablet in India ends up late and doesn't work. Isn't this Just how it always turns out over there?
This is a nice idea, but it flies directly into the extremely strongly held cultural / sociological belief that there is no difference between education and training and they're just synonyms for the same thing. You'd have better luck convincing people God does not exist thru logical argument. "College is training for a good job" is as closely held a belief as "god exists" In some ways, more closely held.
We have badges, they're called certifications, and decades of handing them out like crackerjack prizes means they're mostly useless. Certs with an industry wide accreditation board would work, and to the best of my knowledge has never been tried because of the excuse that developing classes, books, and tests means the cert designers would need to be involved in early blue sky development (as if that were not a pile of B.S. and as if NDAs didn't exist and as if the accreditation board needs to be involved at the earliest level of certification)
Back in 1998 a CCNA meant instant job offer. In 2012 I heard some guys at work talking about getting one because they're a ticket to success. It becomes a belief system, like people who still think getting a law degree means instant millions despite 50% unemployment for law grads for years and years now. If you ever get a badge/cert program in, people will believe in it for decades after it closes down, so maybe a high barrier to entry makes sense after all.
The funny thing is nothing is ever new in IT... Its all the same old stuff with new marketing, as the natural cycle turns. There is a huge competitive on the job advantage in having experienced the previous cycle(s) which noob IT people are completely blind to.
Also its impossible to be a good software engineer or good sysadmin without knowing CS. Almost an oxymoron. Maybe they don't know enough math to follow Knuth, but a good one has at least a gut level instinctual level of low level CS knowledge. Its always hilarious for me to watch IT noobs learn the CS guys not only have formulae and charts and provable theorems about their little scalability problem, but their 2010 gut level guesses and hard fought failed experiments were all figured out by the CS guys in the 60s/70s if they'd only have bothered to learn.
While the line isn't always clear, in general it's NOT OK to lie on resume to obtain a job or gain advancement
I cry bogus. When the economy results in 100 fully qualified applicants for each job, the only way to rise to the top of the resume pile is to lie. Therefore most hired by resume and resume filtration are liars, or at least the percentage of liars is spectacularly high, or honest people are dramatically underemployed. I've gotten all my jobs since 1995 thru "knowing people" and "having heard about me" so I haven't had to lie, I've got no dog in the fight so I can be honest about the situation.
A good resume is a legalistically written document full of unprovable lies.
Seriously, doesn't every resume or cover letter contain BS about being "a team player" and being the cause of incredible business achievements? If all of that crap were true, we'd all be surrounded by superheros, or at least superheros who are mysteriously all retired-in-place. Since we aren't, that would inductively seem to indicate most resumes are works of fiction, at best historically accurate or historically inspired fiction.
Example: I've programmed in Perl for $$ since 1996. I know enough about Perl and CPAN to know its a bigger ecosystem than any human being can really understand in totality even if that is their full time academic job. To me, writing "I know Perl" on my resume feels OK, but I know what it takes to be an expert and I am not an expert because I spend time doing other stuff, other languages, etc. There are a lot of clowns out there who once glanced at a Oreilly book on a shelf and maybe changed a couple strings in a downloaded script who write "I am an expert at Perl" on their resume. One of us is lying, but its a somewhat subjective judgment. To a HR chick, we both know a heck of a lot more Perl than she does, so its technically true from her perspective, and disproving it is going to take an interview by someone with specialized knowledge. The liar sounds like a good interview to schedule to a HR chick. I'll never get that interview from HR because the "expert" resume will rise to the top of the resume pile (well, also ageism means I'm disqualified from any future IT employment by HR, unless I lie on my resume by forward dating graduation years, skipping over a couple jobs, etc... thankfully I don't have gray hair and no wrinkles because I don't suntan). Possibly, if I know the hiring manager and maybe a couple of his underlings, I can bypass HR, get an interview, and get the job despite being "too old" and "not being an expert". That's pretty much how its worked so far.
What is wrong is getting caught by stuffing the resume with stuff that's easily dis-proven. Go ahead, prove I wasn't the NCOIC of my computer control section when I was still in the Army in June of 1995, which is kind of a staffing anomaly for a lowly E-4 (usually a E-6 SSG slot). That story is true, but how the hell you'd prove I was lying is a complete mystery to me. I think they put me there as a test, so both I and them could see if I was going to be a lifer. Answer, a mutually agreed upon no. It was fun, and nothing bad happened, but, no longer interested, thanks. Job assignment should be more like that.
Yes, the CEO is far more important to the company
You can reasonably argue thats not true.
The only constant of corporate life is groupthink concentrates at the top. A standard issue stuffed suit is identical to any other standard issue stuffed suit. Any variation in results by different stuffed suits is caused by natural market variation. Who the board selects is not really all that important. Very much like programmer output, the top 0.1% of rock star workers (programmer or CEO) will outperform the masses by a factor of 10, but there are not enough 0.1% rock stars to really matter.
On the other hand, if you have a peon 1st level cust service who lies all the time, they can destroy millions in shareholder value per week, if they try hard enough.
Not very hard to guess the problem.
Take a look a the third or so pic on http://lab.dereklow.co/brad/ The one with the light switch hanging out of the wall as he screws around with the wiring. If he could have just stuck to plug in modules like everyone else, but no he has to go all amateur electrician... I love this quote "With no access to the circuit breakers of the dorms, the contacts and wiring inside the wall switch remains live even as I open and try to modify it.". This dude is the stereotype of nothing is more dangerous than a programmer with a screwdriver... damn...
Screwing around with the drapes might have pissed them off too. Worst case is losing some security deposit unless he can return it all exactly to original operation.
Oh and the fog machine. My dorm didn't allow gasoline and oil products in the rooms, to discourage people from putting their scooter in the room, or doing oil changes in the dorms. Maybe they're worked up about fog juice and treating it as automotive lubrication oil.
If you're not experiencing the joys of doin' it the hard way, and you don't care if it can't be scaled above 200 million users per site, why not just string together a bunch of CPAN modules and be done in a weekend? That's the part I couldn't figure out.
RoR forces you into certain mindsets for doing things that may make life harder, and the ruby library infrastructure, while good, isn't as wide ranging or as well debugged as CPAN.
I kinda like all 3 languages mentioned, it just seems Perl would have been the correct tool for this particular job.
The funny part is RoR has recently (well, recent YEARS) gone thru the growing pains C++ suffered thru maybe a decade or so ago.
C++ used to be exactly the same way, you upgrade gcc/g++ and suddenly nothing compiles anymore.
I think you're pretty much past that with RoR so its safe?
I think they were also trying to make a political point about scalability. The point of the diaspora is not to centralize, not to scale. There were people who Just Didn't Get It about diasporas goals complaining about RoR "How you gonna scale one RoR site to over 500 million users" etc. Thats, um, kind of the point. Shouldn't have more than a household or so per server. Not enough people per server that the admin could make serious dough selling the private info of people he doesn't care about but are on his server anyway. A self correcting privacy policy, sorta, via dispersal.
Hmm my point was not just fire the bottom half, but to realize that what a smart /.er would consider blindingly obvious might be far beyond the ability of perhaps the bottom half of managers, especially the lower level ones.
Sometimes it sucks to work for a guy who got his job solely based on his golf game, or who his daddy is... that's your fault for not gaming it into a new scenario where it rocks to work for someone who got his job solely based on his golf game, or who his daddy is... In the long run that behavior will destroy the company, but no one cares about the long run anymore, so...
Ah thats boring. /.er you're pretty respectable, but some renters... are not. you can really work a guy across the country...
Wait until it gets exciting, like plumbing leaks, or roof needs replacement.
For extra fun, imagine court case involving the rental property. Or insurance claim (storm damage, etc).
Another good entertainment is local code enforcement.
Finally I'll assume that as a
Now there are property management companies that will take care of these kind of problems... for a fee, often rather high. Your $1600 check is nice, but until you factor in the costs of ownership, prop tax, maintenance can't be deferred forever especially if its rental property, repairs, mortgage if any...
Easier for it to just show up in the mail. I'm very, very, lazy.
Then you're going to be pissed off that several of my reports are now email delivered. No snail mail, or less, anyway.
Examples:
Wisconsin energy is a paper only company. I believe email delivery is optional for quarterlies, but they like to snail mail a big ole paperback book once a year for annual.
I believe PRPFX (a mutual fund) is a email only company. I might have gotten an annual in the snail mail, but I haven't seen a quarterly from those guys in .. forever I guess.
Definitely SLV and GLD (ETFs) are email only companies. I haven't seen paper from those guys since silver was $7 and gold was $500 (which incidentally is pretty close to my DCA share price)
I believe this is a situation where they're legally required to snail mail upon request to a stockholder, so if you really wanted you could send them snail mail and they'd reply with a nice book, in case your wood stove needed kindling. You should make the shareholder services office earn their pay...
I would guess FB will just publish the data to a page and tell people to friend it.