I agree. The kit I had included a pretty powerfull AC motor. One of the projects in the manual was basically a working cattle prod. There's a movie on this sunday, from TV guide:
9:00 PMWCBS Sunday, 15
The Man Who Saved Christmas 120 min.
Jason Alexander stars in this 2002 TV-drama inspired by the true story of toymaker A.C. Gilbert.
In the years before World War I, business isn't booming for Gilbert (Alexander). His father (Edward Asner) wants him to go to medical school, but his wife (Kelly Rowan) and brother (Ari Cohen) persuade him to follow his dream. When A.C.'s Erector Set toy becomes a success, he starts a company that provides a comfortable life for his family. But the Great War comes and so does the Government, asking A.C. to transform his factory into a munitions plant. Not long after, Uncle Sam requests an even greater sacrifice, which threatens Christmas for everyone.
Since the lusers can't remember, then switch to a system that relies on a physical item for the security. Do the employees have ID cards? If so, chances are those could be used for a login system (RFID tags?).
As for explaining to management why passwords are a good idea, ask them if they would like to see their salarys/bonuses/private email show up on F--CKED Company.com (not as a threat, just to point out what can happen when accounts aren't secured without a password or equivilent.)
Unless your buildings are wired by a company like Intellispace, your last mile connection is going to be physically connected by whomever has the local monopoly (like Verizon in NYC). So, if you go with any other company, and installation/troubleshooting needs to be done, you *still* will be dealing with the local monopoly. When they have to prioritze serviceing one of their customers, or a 3rd party's customer, guess who wins? 4 months of T-1 hell made that point very clear to me.
if you go to http://www.shrevesystems.com/powerbook.html you'll get their old site, but no way to order online. I think there's an onsite auction next week. And Shreveaudio is/was part of the computer shreve.
Buckyballs weren't found in the wild until after they were made in a lab. My only point is that sometimes things in the wild aren't found because we haven't been looking specifically for it.
WinImages
Yes, I know this is a windows program, so get Virtual PC to run it. There's a sale on WinImages, it's now $50. This is one of the best morphing programs, it's scriptable, and does a hundred other effects besides morphing. And since this is slashdot, I'll need to mention that their credits page includes X-Men
THANK YOU! This fits nicely with the previous/. discussion on solid state drives. With motherboards supporting 3 or 4 gigs ram, a virtual ram drive could be a great thing to use as the primary pshop/gimp scratch drive.
Is the data changing where you will have hundreds of gigs being replaced each day? Or is this a write once read many application? If it is mainly a WORM then what about using proms? Or use a form of optical storage that will have a minimum of moving parts, perhaps using crystals instead of rotating mirrors or media.
umnn, it's even worse than that. The MTBF is calculated based on REPLACEING the drive before its end of service life. So drives may actually need to be replaced every year to make that series of drives get its million hour MTBF here is a good article on MTBF
also check out www.wavecorp.com's MediaBank. If they've cleared up some of the bugs left over from previous owners, it may be a good contender. The version control was nice. A few years ago their only backup system was Legato, which was a bit kludgey (for example, Legato can return an error that the archive DID NOT complete sucessfully, which MediaBank ignored, so it did its default action of deleting what had been archived)
I've heard very good things about the Xinet solution. They're a great company to work with, very responsive to problems. (which are rare)
One important thing about DAM is that the users *WILL NOT* voluntairily change their workflow, and will not follow guidelines on folder naming, adding notes etc. Even if the new way with the DAM is better/faster, a lot of users won't want to switch. You have to come up with ways that make it difficult to bypass your system. For example, users will move or add/delete files using the Finder rather than the DAM, causing the DAM to lose sync with the filesystem. Mediabank had the option to access files directly thru TCP/IP, so that the server volumes did not have to be mounted.
Lastly, do a beta test. Also try some disaster senarios. Dead Drives, viruses deleting jpgs, corrupt database, drives filling to 100%, loss of ethernet switch while client has open files, etc. Yes, I've had to deal with all of these:-(
Re:whoever wrote this artlcle is on crack.
on
Open Blade Servers?
·
· Score: 1
>>That bundle includes a lot more than just 5 blades.
well, yeh, but that doesn't change that point since you need the chassis to run the blades. I did check to make sure that the config of a stand alone was better than the config of the special, so it was an apples to apples comparison.
>>Economies of scale quickly kick in once you've bought a chassis (like the 3U) or two.
ditto for buying pc's in bulk. What is the point when the blade costs (blades + chassis needed) matches stand alone?
I *like* the idea of a blade system. Unfortunatly, since RLX doesn't have the volumes that standard motherboards have, you probably don't have the economies of scale that they have
Re:whoever wrote this artlcle is on crack.
on
Open Blade Servers?
·
· Score: 1
I just looked at the RLX site, and according to the website promo 5 blades will cost over $10K. And that's just for the midrange version. Since chassie price on the 24 blade version wasn't listed I don't know its cost.
My real point is that if we were looking at buying 100 servers a price difference of $100 over other systems would be sellable, but if the difference is over $800 (similar shuttle system, better config)the beancounters may put some of that $80K savings toward another rack or two......
Might be an idea for RLX to make a chassie that can use standard motherboards (ITX?)and 3.5 disks.
>>If you want a full 1.5 Mbps internet conection, you must pay from $700 to $1500 for a T1, depending on location.
Based on a few T-1's I've ordered in NYC, about half the cost is FCC fees and taxes, which I'm pretty sure aren't charged for DSL and cable. The more expensive CSU/DSU and quality level and phone line quality are other expenses that don't occur with a cable connection.Plus, if it's the same company that owns some backbone, costs should be cheaper.
Additionally a google search yielded several co-lo's that charge about $2/gig for bandwidth in bulk. Another trick that TW/RR does is making the cost of standard cable + Road Runner be just a little more than just cable alone. So with one wire they get 2 revenue streams.
Basically, I *think* that their actual cost per user is around $5/month, plus a buck a gig for transfer. Is there anyone out there who works for a cable company who can confirm or correct this?
Since the cable company has a goverment granted monopoly in most cities, citizen/consumers should have a greater say in pricing than if it were a truely free market. And yes, it does run the risk of a California power supply style crisis happening to cable based ISP's, so some form of controls need to be in place so that we do end up paying for fair market value bandwidth usage.
Just like Microsofts infamous benchmarks, the data about "bandwidth hogs" and "mom&pop" usage is being supplied by the cable companies. I think we need an open source project to gather a couple of months worth of data on actual usage.
I'd imagine it could work like this: The client could be placed not only on our machines, but also on friends and family's machines so that "normal" users would be recorded. Basic info like tested speeds, area, type of connection would be recorded, but not the type of info that may cause accusations that it's a marketing scam. At the start of each new internet session (or once a day) it would send usage data to two or more servers. The raw data would be accessable to anyone so that any conclusions or statistics derived could be verified. The main problem I see is preventing false data from being submitted to the servers.
Wouldn't it be nice to show up at a hearing with proof that the companies are lying to the committe? Of course the flip side is that this project may definitivly prove that the cable companies are correct, and that slashdotians are bandwith hogs.....
but his employer IS NOT paying him. He said he is a part owner, so he essentially employes himself. That $1200 is $1200 lost profit, part of which may have gone to him as cash. Basically, he paid for the books himself.
http://www.upshotstock.com/Default.htm this may be what you are looking for.
I just found this by going to http://directory.google.com/Top/Business/Arts_and_ Entertainment/Photography/Stock/
Almost every agency only deals with collections (100+ images) from photographers, since dealing with tens of images from hundreds of photogs would be a nightmare of maintaining a consistant quality. If there is a problem with the image, the website would be blamed, not the photog who messed up.
About 4 years or so ago I worked on a system for a major NYC lab. It was capable of quickly making a skinable site for any photographer who wanted to sell their (royalty free) images online.
The biggest problem was not DRM, but simply that the larger size previews were perfect for webuse. We ended up using a embossed watermark on the preview image.
For the collections, we usually did a 60 meg scan, archived as a tiff onto a nearline DLT robot. For web distribution the dpi was cut in half, flashpix with jpeg, so the dialup users only had to deal with 4 meg files.
If I were to do it today the originals would be saved in a wavelet (mr. sid, lurawave, jpeg2K etc) at a high quality setting, stored online. It would then be pretty simple to give the user any size/crop file they wanted.
Thanks for the question, since I hadn't looked at what's out there for a few years
Untill you've actually been an owner of a small business, you really can't appreciate the point of view of "the pointy haired ones".
Since most small business fail, and you'll probably disolve it when you do go to college, assume that the money you invest in it will be lost.
You'll get to learn about corporations, fed/state/local taxes, 1099 forms and reporting, dealing with irate customers who owe you money, dealing with employees who owe you work, double entry accounting and deadlines that if missed would be a lot more serious than turning in a book report late.
I was a partner in a small multimedia company that lasted 7 years and I am still amused by the misperceptions of people who've never run a business. for example, a typical conversation with some friends and relatives: "why don't they give you a raise?" "they is me" "so you can give yourself a raise?" "sure, if I want to meet some nice repro men when they cart away my office furniture for missed lease payments" "but you work such long hours, shouldn't you get more money" "yes, I should. And if a client didn't go chapter 7 oweing $20K, I might be able to" "but what does that have to do with getting a raise?"
There's lots of resources available, maybe start by watching Rodney Dangerfield's "Back to School"
>>"Why shouldn't people be able to simply type (or better yet , speak) the command 'turn the text in the box red if the amount is less than zero' instead of the incomprensible code"
...are doomed to repeat it" Lets hope they have learned something from the book "The Hubble Wars: Astrophysics Meets Astropolitics in the Two-Billion-Dollar Struggle over the Hubble Space Telescope" by Eric J. Chaisson
Build a device that would house the message on a banner, have a timer and anti-tampering systems.
Hide the devices in populated areas where it won't be noticed. For example; the backs of highway signs, roofs of buildings, subway tunnels, highway medians, bottom of ponds, above the ceiling tiles in a library, etc.
On the day of revilation, the banners pop out.
or, let Verisign help. Get a few websites prepaid for a few years. Get a few domain names. Encrypt the discovery using an ip address or three as the key. On the release day(s) a few of the websites submit a change of ip command to verisign. Then the sites with the info can get the key by resolving the domain name.
Simple. Sign a reseller agreement that defines who, what and where you can sell the products. This is/was pretty common, and a good thing, because as a reseller you didn't have to worry much about out of town competition, or someone who had an only educational sales agreement wouldn't be able to undercut your comercial sales agreement.
Wow, in a twisted way it's nice to see other people who hate that guy. I feel less alone:-)
And it's really ironic AC said the founder likes to sue people since he was lucky I didn't sue him when he screwed me over. Since you may "enjoy" this, and it's a good forum to rant and maintain grudges, here's the Off Topic story.
Back when Truespace was around 1.x on the Amiga, I was a partner in a fledgling multimedia company. We got Bantam as a client, and got the job to render 4 versions of their logo embossed in gold on marble cubes, to be used as a printed cover, so a 2k image was needed. At that time Caligari was running ads claiming that it was capable of rendering 3K images suiteable for film output
Since we had a good relationship with Amagination (amiga store/multimedia) we outsourced the modeling/rendering to them. After a difficult modeling job, and over 30 hours of rendering, with our deadline looming, we had a 2k file suiteable for film output Since the file was in caligari's format, it had to be converted into targa. or tiff. or bmp. or pict, or any format that some other program could read. Seems that there was just one third party utility that could read and convert it, and caligari had never tested it beyound screen res. There was a bug, and anything over 512 pixels tall couldn't be converted. Fortunatly, "that guy" was located in NYC, so i brought the files (output and models/textures) to him for debugging. At first he tried to hide behind that fact that since his file was 3k it would be good for film output. It took awhile to make the point that if we had no way to start with his file, and end up with a physical piece of film in my hand, then his ad was a lie, and he should fix it now.
After a week or so we gave up waiting for him to fix it, and went to another company to do it in Topaz. This caused weeks of delay, and we lost thousands of dollars. Very painfull, and we never got anymore work from Bantam again.
What really rubbed salt in the wound was a few weeks later at an Amiga show, Roman was demoing using OUR FILES to show how wonderfull calagari was. Umnnn, do ya think he had asked permission to use confidential copyrighted files in a public demo? When I finally stopped screaming at him, he may have gotten the idea that we weren't happy with what he was doing. Fortunatly the security guards were persuaded not to throw me out. Not having deep pockets (hell, at that time we couldn't afford pockets) we didn't sue, and just hoped bad Karma would do something. The real shame of this was that calagari really did have beautifull renders/lighting.
Slashdot Mirror
There's a movie on this sunday, from TV guide:
9:00 PMWCBS Sunday, 15
The Man Who Saved Christmas
120 min.
Jason Alexander stars in this 2002 TV-drama inspired by the true story of toymaker A.C. Gilbert.
In the years before World War I, business isn't booming for Gilbert (Alexander). His father (Edward Asner) wants him to go to medical school, but his wife (Kelly Rowan) and brother (Ari Cohen) persuade him to follow his dream. When A.C.'s Erector Set toy becomes a success, he starts a company that provides a comfortable life for his family. But the Great War comes and so does the Government, asking A.C. to transform his factory into a munitions plant. Not long after, Uncle Sam requests an even greater sacrifice, which threatens Christmas for everyone.
Since the lusers can't remember, then switch to a system that relies on a physical item for the security. Do the employees have ID cards? If so, chances are those could be used for a login system (RFID tags?).
As for explaining to management why passwords are a good idea, ask them if they would like to see their salarys/bonuses/private email show up on F--CKED Company.com (not as a threat, just to point out what can happen when accounts aren't secured without a password or equivilent.)
Unless your buildings are wired by a company like Intellispace, your last mile connection is going to be physically connected by whomever has the local monopoly (like Verizon in NYC).
So, if you go with any other company, and installation/troubleshooting needs to be done, you *still* will be dealing with the local monopoly. When they have to prioritze serviceing one of their customers, or a 3rd party's customer, guess who wins? 4 months of T-1 hell made that point very clear to me.
if you go to http://www.shrevesystems.com/powerbook.html you'll get their old site, but no way to order online. I think there's an onsite auction next week. And Shreveaudio is /was part of the computer shreve.
Buckyballs weren't found in the wild until after they were made in a lab. My only point is that sometimes things in the wild aren't found because we haven't been looking specifically for it.
WinImages
Yes, I know this is a windows program, so get Virtual PC to run it. There's a sale on WinImages, it's now $50. This is one of the best morphing programs, it's scriptable, and does a hundred other effects besides morphing. And since this is slashdot, I'll need to mention that their credits page includes X-Men
Looks like nothing has changed since this was previously asked.
Devouring Fungus: Tales of the Computer Age by Karla Jennings
serial ATA is the next standard for drives.
THANK YOU! /. discussion on solid state drives. With motherboards supporting 3 or 4 gigs ram, a virtual ram drive could be a great thing to use as the primary pshop/gimp scratch drive.
This fits nicely with the previous
Is the data changing where you will have hundreds of gigs being replaced each day? Or is this a write once read many application? If it is mainly a WORM then what about using proms? Or use a form of optical storage that will have a minimum of moving parts, perhaps using crystals instead of rotating mirrors or media.
umnn, it's even worse than that. The MTBF is calculated based on REPLACEING the drive before its end of service life. So drives may actually need to be replaced every year to make that series of drives get its million hour MTBF
here is a good article on MTBF
also check out www.wavecorp.com's MediaBank.
:-(
If they've cleared up some of the bugs left over from previous owners, it may be a good contender. The version control was nice. A few years ago their only backup system was Legato, which was a bit kludgey (for example, Legato can return an error that the archive DID NOT complete sucessfully, which MediaBank ignored, so it did its default action of deleting what had been archived)
I've heard very good things about the Xinet solution. They're a great company to work with, very responsive to problems. (which are rare)
another one to check is http://www.artesia.com/teams_overview.html
Why do you dislike MetaCommunications?
One important thing about DAM is that the users *WILL NOT* voluntairily change their workflow, and will not follow guidelines on folder naming, adding notes etc. Even if the new way with the DAM is better/faster, a lot of users won't want to switch. You have to come up with ways that make it difficult to bypass your system. For example, users will move or add/delete files using the Finder rather than the DAM, causing the DAM to lose sync with the filesystem. Mediabank had the option to access files directly thru TCP/IP, so that the server volumes did not have to be mounted.
Lastly, do a beta test. Also try some disaster senarios. Dead Drives, viruses deleting jpgs, corrupt database, drives filling to 100%, loss of ethernet switch while client has open files, etc. Yes, I've had to deal with all of these
>>That bundle includes a lot more than just 5 blades.
well, yeh, but that doesn't change that point since you need the chassis to run the blades. I did check to make sure that the config of a stand alone was better than the config of the special, so it was an apples to apples comparison.
>>Economies of scale quickly kick in once you've bought a chassis (like the 3U) or two.
ditto for buying pc's in bulk. What is the point when the blade costs (blades + chassis needed) matches stand alone?
I *like* the idea of a blade system. Unfortunatly, since RLX doesn't have the volumes that standard motherboards have, you probably don't have the economies of scale that they have
My real point is that if we were looking at buying 100 servers a price difference of $100 over other systems would be sellable, but if the difference is over $800 (similar shuttle system, better config)the beancounters may put some of that $80K savings toward another rack or two......
Might be an idea for RLX to make a chassie that can use standard motherboards (ITX?)and 3.5 disks.
>>If you want a full 1.5 Mbps internet conection, you must pay from $700 to $1500 for a T1, depending on location.
Based on a few T-1's I've ordered in NYC, about half the cost is FCC fees and taxes, which I'm pretty sure aren't charged for DSL and cable. The more expensive CSU/DSU and quality level and phone line quality are other expenses that don't occur with a cable connection.Plus, if it's the same company that owns some backbone, costs should be cheaper.
Additionally a google search yielded several co-lo's that charge about $2/gig for bandwidth in bulk.
Another trick that TW/RR does is making the cost of standard cable + Road Runner be just a little more than just cable alone. So with one wire they get 2 revenue streams.
Basically, I *think* that their actual cost per user is around $5/month, plus a buck a gig for transfer. Is there anyone out there who works for a cable company who can confirm or correct this?
Since the cable company has a goverment granted monopoly in most cities, citizen/consumers should have a greater say in pricing than if it were a truely free market. And yes, it does run the risk of a California power supply style crisis happening to cable based ISP's, so some form of controls need to be in place so that we do end up paying for fair market value bandwidth usage.
Just like Microsofts infamous benchmarks, the data about "bandwidth hogs" and "mom&pop" usage is being supplied by the cable companies.
I think we need an open source project to gather a couple of months worth of data on actual usage.
I'd imagine it could work like this:
The client could be placed not only on our machines, but also on friends and family's machines so that "normal" users would be recorded. Basic info like tested speeds, area, type of connection would be recorded, but not the type of info that may cause accusations that it's a marketing scam. At the start of each new internet session (or once a day) it would send usage data to two or more servers. The raw data would be accessable to anyone so that any conclusions or statistics derived could be verified.
The main problem I see is preventing false data from being submitted to the servers.
Wouldn't it be nice to show up at a hearing with proof that the companies are lying to the committe? Of course the flip side is that this project may definitivly prove that the cable companies are correct, and that slashdotians are bandwith hogs.....
but his employer IS NOT paying him. He said he is a part owner, so he essentially employes himself.
That $1200 is $1200 lost profit, part of which may have gone to him as cash. Basically, he paid for the books himself.
http://www.upshotstock.com/Default.htm
_ Entertainment/Photography/Stock/
this may be what you are looking for.
I just found this by going to http://directory.google.com/Top/Business/Arts_and
Almost every agency only deals with collections (100+ images) from photographers, since dealing with tens of images from hundreds of photogs would be a nightmare of maintaining a consistant quality. If there is a problem with the image, the website would be blamed, not the photog who messed up.
About 4 years or so ago I worked on a system for a major NYC lab. It was capable of quickly making a skinable site for any photographer who wanted to sell their (royalty free) images online.
The biggest problem was not DRM, but simply that the larger size previews were perfect for webuse. We ended up using a embossed watermark on the preview image.
For the collections, we usually did a 60 meg scan, archived as a tiff onto a nearline DLT robot. For web distribution the dpi was cut in half, flashpix with jpeg, so the dialup users only had to deal with 4 meg files.
If I were to do it today the originals would be saved in a wavelet (mr. sid, lurawave, jpeg2K etc) at a high quality setting, stored online. It would then be pretty simple to give the user any size/crop file they wanted.
Thanks for the question, since I hadn't looked at what's out there for a few years
Untill you've actually been an owner of a small business, you really can't appreciate the point of view of "the pointy haired ones".
Since most small business fail, and you'll probably disolve it when you do go to college, assume that the money you invest in it will be lost.
You'll get to learn about corporations, fed/state/local taxes, 1099 forms and reporting, dealing with irate customers who owe you money, dealing with employees who owe you work, double entry accounting and deadlines that if missed would be a lot more serious than turning in a book report late.
I was a partner in a small multimedia company that lasted 7 years and I am still amused by the misperceptions of people who've never run a business. for example, a typical conversation with some friends and relatives:
"why don't they give you a raise?"
"they is me"
"so you can give yourself a raise?"
"sure, if I want to meet some nice repro men when they cart away my office furniture for missed lease payments"
"but you work such long hours, shouldn't you get more money"
"yes, I should. And if a client didn't go chapter 7 oweing $20K, I might be able to"
"but what does that have to do with getting a raise?"
There's lots of resources available, maybe start by watching Rodney Dangerfield's "Back to School"
>>"Why shouldn't people be able to simply type (or better yet , speak) the command 'turn the text in the box red if the amount is less than zero' instead of the incomprensible code"
Hmnn, that's almost a legal line of applescript
...are doomed to repeat it"
Lets hope they have learned something from the book "The Hubble Wars: Astrophysics Meets Astropolitics in the Two-Billion-Dollar Struggle over the Hubble Space Telescope"
by Eric J. Chaisson
Build a device that would house the message on a banner, have a timer and anti-tampering systems.
Hide the devices in populated areas where it won't be noticed. For example; the backs of highway signs, roofs of buildings, subway tunnels, highway medians, bottom of ponds, above the ceiling tiles in a library, etc.
On the day of revilation, the banners pop out.
or, let Verisign help.
Get a few websites prepaid for a few years. Get a few domain names. Encrypt the discovery using an ip address or three as the key. On the release day(s) a few of the websites submit a change of ip command to verisign. Then the sites with the info can get the key by resolving the domain name.
Simple. Sign a reseller agreement that defines who, what and where you can sell the products. This is/was pretty common, and a good thing, because as a reseller you didn't have to worry much about out of town competition, or someone who had an only educational sales agreement wouldn't be able to undercut your comercial sales agreement.
Since you may "enjoy" this, and it's a good forum to rant and maintain grudges, here's the Off Topic story.
Back when Truespace was around 1.x on the Amiga, I was a partner in a fledgling multimedia company. We got Bantam as a client, and got the job to render 4 versions of their logo embossed in gold on marble cubes, to be used as a printed cover, so a 2k image was needed.
At that time Caligari was running ads claiming that it was capable of rendering 3K images suiteable for film output
Since we had a good relationship with Amagination (amiga store/multimedia) we outsourced the modeling/rendering to them.
After a difficult modeling job, and over 30 hours of rendering, with our deadline looming, we had a 2k file suiteable for film output
Since the file was in caligari's format, it had to be converted into targa. or tiff. or bmp. or pict, or any format that some other program could read. Seems that there was just one third party utility that could read and convert it, and caligari had never tested it beyound screen res. There was a bug, and anything over 512 pixels tall couldn't be converted. Fortunatly, "that guy" was located in NYC, so i brought the files (output and models/textures) to him for debugging.
At first he tried to hide behind that fact that since his file was 3k it would be good for film output. It took awhile to make the point that if we had no way to start with his file, and end up with a physical piece of film in my hand, then his ad was a lie, and he should fix it now.
After a week or so we gave up waiting for him to fix it, and went to another company to do it in Topaz. This caused weeks of delay, and we lost thousands of dollars. Very painfull, and we never got anymore work from Bantam again.
What really rubbed salt in the wound was a few weeks later at an Amiga show, Roman was demoing using OUR FILES to show how wonderfull calagari was. Umnnn, do ya think he had asked permission to use confidential copyrighted files in a public demo? When I finally stopped screaming at him, he may have gotten the idea that we weren't happy with what he was doing. Fortunatly the security guards were persuaded not to throw me out. Not having deep pockets (hell, at that time we couldn't afford pockets) we didn't sue, and just hoped bad Karma would do something. The real shame of this was that calagari really did have beautifull renders/lighting.