It may be a good insider joke, but I'm sure people who've actually sunk money into this "company" are far from amused. (No, I don't play stocks.)
I'd seriously like to know how many years of nonsense marketing, nonexistent products, and continued market blather are required before an investor group or the government finally says "Enough!" and shuts them down or jails those responsible.
Partition drive. Install WinXP to C:. Use boot disk to install Linux on another partition/drive, and let it install GRUB.
No problems in 5+ years with McAfee, Norton, F-Secure, or any other virus protections I've used.
The one catch is you need to disable BIOS MBR protection during your installs and updates, but as I've never heard of anyone stopping a virus via BIOS "anti-virus" support, I just leave the "protection" disabled.
With the exception of ADA and maybe Modula-2, J2SE5 is one of the most type-safe languages I've used.
Java's "typecasting" is not like a C-cast, but rather like a full C++ dynamic_cast with RTTI enabled.
There is a big difference between type-safe runtimes and compile time type checking.
Where pretty much every language other than ADA falls down is their lack of numeric precision enforcement -- it's almost impossible to implement correct calculations in C/C++ without unwinding the statements so you can do correct result truncation after each operation involved. Languages which do such precision checking and enforcement don't suffer when automatic type conversions are performed, because you get told about the overflow/underflow issues.
Don't think for a minute that the C++ "void *" is anything like Java's "Object". The former is just an address/pointer, the latter carries type information.
On a production box, the admins have access to sudo, and root itself is locked down except for scheduled maintenance/upgrades or emergencies. No paperwork, no root.
As a developer with over 15 years *nix experience, I have never had root access to a box unless I was doing an install, except for my own desktop workstation. In the case of my desktop, the only reason developers had root was so we could kill rogue services during debug sessions gone bad.
Under no circumstances do I agree with any user installing additional software on a box. If it's needed, it gets approved and installed for everyone who needs the functionality, not by rogue users.
I think the original poster did the right thing by giving notice.
Personally I've never had my system access shut off because I gave notice, but I did have it shut off before I was told I was laid off (many years ago.) However, as I've worked in a number of verticals I know there are a few that disable access to live systems, but let developers keep working during their notice period.
With live/production data, there are often regulations that would prohibit allowing a sysadmin to continue accessing the system after they've given notice. I realize it probably feels insulting to have your access shut off after acting like a professional and giving notice, but I wouldn't take it personally.
Reads like a classic project failure, with the classic failed project start: It was managed by someone who created "their own" database. i.e. A manager who thinks he knows better than the experts being hired, who overrides their estimates and recommendations, and who blows off any technical issues they raise because he "did it himself" in less time with an underpowered single-user tool.
I've worked on three similar projects -- only one succeeded. The one success was because the manager in question got yanked half way through and there was still enough time for a competent manager to be hired who'd let the team do their job properly.
I can think of a few other potentials, like the fabric of a chair sensing where the pressure points are and adjusting itself accordingly. I could see combining such a pressure-sense technology with a body-height cabinet for taking full-body measurements for custom clothing manufacture, and even more mundane uses like a more accurate "touch" scoring system for medieval recreationists.
For general robotics systems I can see more safety uses than practical -- wrap the limbs of the robot in this stuff so it can detect a collision that is out of it's field of view. (You just know that if we ever have household robots the kids are gonna climb on the things and one of the kids will end up hurt because the robot didn't see them.)
It looks like I'll have to do a bit of reading up on AI systems some day. As far as I knew, the issue was more one of neural learning networks vs. expert systems approaches.
Learning networks acquire and test connections and boundary conditions, keeping those which seem to be most relevant to the "world" as the AI learns. Expert systems are more predictable, but they only capture human knowledge for automation rather than learning new approaches.
From what you're saying, it sounds like the AI community breaks things down much farther. Personally I wouldn't have thought of constraint logicistics as being any different from tree pruning for an inference engine. Maybe algorithmically more efficient for certain classes of problems, but not functionally different.
Don't count on it. One of the famous government wastes in western Canada was a multi-year study on cow manure. Seriously. Not methane production, but manure. *shrug*
But I could imagine this maybe having impact on related fields that work with pressure-molded liquids or semi-solids. I'd think there are similar issues with producing an even coating/coloring for the glass part of a light bulb, for example. Sure it's not that they don't have ways of coating a bulb, but maybe his work might apply to a better way of embedding what is currently a coating, or applying it more evenly, or allowing the use of thinner glass, etc.
Not that I necessarily think it was worth 11 years, but having spent more than that on my own pet project I can well understand someone getting a tad fanatical about something no one else necessarily sees a use for.:)
The one I like is "Nobody will want to pull a Sony now."
Like it's a bad thing that corporations learn that invasive and damaging approaches to IP protection actually tick off consumers and cause lawsuits? The only surprise is that they had to learn by doing instead of just asking a few computer maintenance techs how they feel about the rootkit approach!
This was roughly 8 years ago. The town is north-west of Brampton/Bramalea.
It is no misinformation -- a number of techies I knew had planned on buying houses there until they found out the local cableco had been blocked from delivering high speed by the town's bylaws.
Most likely the bylaw has been fought and dumped since then, but the fact that it ever passed just goes to show the power of paranoid voters.
There is a suburb of Toronto that actually made it illegal for anyone to provide high-speed connections. IIRC, 64K is the limit because the "citizens" don't want their town to become a bedroom community of telecommuting techies.
Insane as it sounds, the town successfully passed the law. I don't know if it was ever overturned as an infringement of rights and freedoms, but the fact that a town could pass such an insane law means that a country could do the same.
Never underestimate the voting power of the retirees, blue collars, and low income people who hate the internet and even blame it (and computers in general) for the loss of jobs and the general dissolution of society.
Personally I find it amusing that management places so much emphasis on buzzwords like RAD, Scrum, and Extreme Programming. All most of these methodologies really do is emphasize what was known back in the sixties: a few skilled developers can code rings around a large team bogged down by communications and paperwork.
While both QT and GTK+ applications cross-compile to various *nix distributions and Windows, GTK+ does not require licensing fees to produce a closed-source application like Acrobat. While there is a "free" version of QT for Windows, it's only free for GPL applications IIRC. You can't even use it for development of a non-GPL application and defer the licensing fees until you have been paid by a client.
Another useful feature of Gnome is it's underlying reliance on CORBA technology. There are CORBA gateways for J2EE (and presumably.NET) that should allow you to build a native client against a TP-based J2EE server with Gnome. I don't believe you can do that with KDE/QT, at least not as tightly integrated/standardized as with Gnome. I don't know of any companies that have actually implemented such an application, but it would theoretically allow you to address the GUI performance issues while still maintaining a J2EE infrastructure that can also support web interfaces for external users.
Ok, now this is getting bizarre. This week's episode of ReGenesis on CTV dealt with the old Spanish Flu virus being recovered from a corpse frozen in the permafrost of Nunavut. Either ReGenesis was dealing with very current information when it was scripted, the science teams are doing "No, it's really true!" reporting after the episodes, or someone isn't doing their fact checking.
Google could just ensure that their test team is testing major vendor's hardware like Dell, HP, etc. After all, if you're talking about business and joe user functionality, you don't need to focus on 3D acceleration and such.
Google could just sink their cash into Novell/SuSE, RedHat, or Mandriva and provide a bundle that already works. Oh, wait, that's right -- you can already get Linux bundles with Java, OO/SO, etc.
So what's the "new" aspect you're suggesting, other than Google becoming involved in the marketing and distribution? What precisely is it that we need for a desktop GUI that isn't already in KDE and/or Gnome? 3D alpha-transparency spinners? Corona effects for the "glint" off metallic 3D lettering?
What Google could really provide in this area is some funding to improve the hardware support and configuration/maintenance utilities for components like configuring 3D support, adding/removing software, etc. I'm not talking about yet another front-end for RPM or APT, but some real improvement in reducing dependencies and manageability.
It's a way of implementing thin clients, but I don't know that it's a "crappy" way of doing so. The alternatives are facilities like X-terminals with an application server or equivalent services for Windows. As a browser-based thin client at least offloads a lot of the display updates to client-only traffic instead of network traffic, it would make a rather substantial difference in the bandwidth required by a thin client to edit documents.
Besides, don't confuse "potentially useful" with "good idea." It's just an interesting approach, and it seems to work ok for very simple documents like email text and web forum editors.
Linus is talking about hardware specs, then comparing them to mixed model specs like OSI. He further generalizes that specs are useful for talking about implementations, but not for actually doing them.
The problem is that he's ignoring how important that discussion framework is when trying to bring new people into a project, or trying to maintain components after the lead engineer(s) are gone. A good spec provides an invaluable model of how to think about the implementation, not a written-in-stone detail of how to write the implementation.
But he is right about bad specs. I once worked with some old Sperry mil-spec hardware (AN-UYK502.) The compilers produced code based on the hardware spec, but the real hardware implementation had reversed the registers for the results of division. Every piece of code produced by the compiler had to be manually tweaked to correct the mistake, but because the compiler met the spec, the vendor wasn't allowed to fix the problem. To the military mindset, it was the hardware that didn't meet spec and needed to be fixed.
Reliability is another issue. If I have the documents on my local drive, I can always retrieve them from backups if the system goes down. But if the internet link is down, it doesn't matter that I've got other computers available -- I still can't get to my documents.
Where I do see such technology having potential use is for the intranet, where it's already common for documents to be stored on a central server instead of local hard drives.
Microsoft, *IAA, etc. know they can't win against offshore firms and open source under the current global legal system. They're pushing hard to have US laws (and presumably the US patent/copyright databases) applied globally.
The manager claims to have run SAP on AIX for several years, and to have been comfortable with that. Given the minimal pricing difference between Linux/x86 and AIX/POWER for IBM bundles, why wouldn't they have gone with AIX in the first place?
Don't get me wrong -- I've found Linux to be very solid for a number of years. But when you're dealing with package compatabilities and interdependencies, sometimes a commercial OS is easier to track and maintain.
It's also not very clear whether the core problem is with SAP or RedHat. Given the engineering squads that are required to customize an SAP installation, I'd really hesitate to presume the problems couldn't have been faulty customization rather than either of the core products.
The only so-called "database" that emphasizes it's GUI is Access. Every other vendor/product I'm aware of relies of separation of duties and doesn't try to roll user interfacing into what is rightly a back-end service.
Administration tools for commercial and OSS databases may be easy for small sites and novice DBA's that don't know their tools, but large applications rely on database scripts to handle configuration, not GUIs. The reason is simple: you can't put a mouse click into CVS/RCS/SCCS/???.
All laptops, workstations, and PCs provided by the company are the property of the company. They are provided to help you do your job.
Password protect all accounts. Enforce password hardening and mandate changing passwords every 90 days.
Secure the root, Administrator, and service accounts. General users should not have access to these accounts.
Ensure that all company approved and installed software is updated in a timely fashion. Have an emergency rollout plan in place for high priority security fixes.
Maintain a corporate update server for deploying software updates.
Check systems for unauthorized software on a weekly basis.
Identify and deploy anti virus products for required client platforms.
Schedule weekly full system scans of all client workstations and PCs.
Do not allow laptops, PCs, or workstations to be connected to the internal networks without updating the anti virus.
Do not allow laptops, PCs, or workstations to connect to internal networks if full system scan has not been performed in the past seven days.
Identify and deploy anti spyware products for required client platforms.
Schedule weekly full system scans of all client workstations and PCs.
Do not allow laptops, PCs, or workstations to be connected to the internal networks without updating the anti spyware.
Do not allow laptops, PCs, or workstations to connect to internal networks if full system scan has not been performed in the past seven days.
Identify and deploy firewall products for required client platforms.
Default deny all incoming ports.
Servers open specific incoming ports required to publish and run their service.
Require management signoff to forward service ports to the public internet.
Assign all users a corporate email account id.
Default-deny internet email for user accounts, restricting them to emailing in the company domain.
Management signoff is required if a task/role requires internet email.
Run email through as many virus scanning tools as you can afford.
Identify supported email clients for required platforms. Ensure that email clients are configured to virus scan all incoming emails.
Quarantine all emails with executable, screen saver, or other high-risk attachments.
Secure all internet access.
Use password-protected proxy servers to prevent unaudited use of corporate resources.
Audit access to external websites and ensure that all users are aware of the policy.
Flag all attempts to access questionable sites.
Block all attempts to access blacklisted sites.
Install and configure office automation software. Management signoff is required.
Install and configure additional software as per management signoffs.
IIIR has no website which can be located by Google, Yahoo, or any other search engine that I've used.
All that comes up are some investor reports like OneSource, which reports them as having a whole seven employees.
Trademarks are not automatically international, and the mere presence of the "G" before "Mail" is not a trademark. Trademarks are either specific spellings (with/without hyphens), logos, icons, color combinations, etc.
The lawsuit sounds like basic "trolling for dollars", legal style.
Slashdot Mirror
It may be a good insider joke, but I'm sure people who've actually sunk money into this "company" are far from amused. (No, I don't play stocks.)
I'd seriously like to know how many years of nonsense marketing, nonexistent products, and continued market blather are required before an investor group or the government finally says "Enough!" and shuts them down or jails those responsible.
Partition drive. Install WinXP to C:. Use boot disk to install Linux on another partition/drive, and let it install GRUB.
No problems in 5+ years with McAfee, Norton, F-Secure, or any other virus protections I've used.
The one catch is you need to disable BIOS MBR protection during your installs and updates, but as I've never heard of anyone stopping a virus via BIOS "anti-virus" support, I just leave the "protection" disabled.
With the exception of ADA and maybe Modula-2, J2SE5 is one of the most type-safe languages I've used.
Java's "typecasting" is not like a C-cast, but rather like a full C++ dynamic_cast with RTTI enabled.
There is a big difference between type-safe runtimes and compile time type checking.
Where pretty much every language other than ADA falls down is their lack of numeric precision enforcement -- it's almost impossible to implement correct calculations in C/C++ without unwinding the statements so you can do correct result truncation after each operation involved. Languages which do such precision checking and enforcement don't suffer when automatic type conversions are performed, because you get told about the overflow/underflow issues.
Don't think for a minute that the C++ "void *" is anything like Java's "Object". The former is just an address/pointer, the latter carries type information.
On a production box, the admins have access to sudo, and root itself is locked down except for scheduled maintenance/upgrades or emergencies. No paperwork, no root.
As a developer with over 15 years *nix experience, I have never had root access to a box unless I was doing an install, except for my own desktop workstation. In the case of my desktop, the only reason developers had root was so we could kill rogue services during debug sessions gone bad.
Under no circumstances do I agree with any user installing additional software on a box. If it's needed, it gets approved and installed for everyone who needs the functionality, not by rogue users.
I think the original poster did the right thing by giving notice.
Personally I've never had my system access shut off because I gave notice, but I did have it shut off before I was told I was laid off (many years ago.) However, as I've worked in a number of verticals I know there are a few that disable access to live systems, but let developers keep working during their notice period.
With live/production data, there are often regulations that would prohibit allowing a sysadmin to continue accessing the system after they've given notice. I realize it probably feels insulting to have your access shut off after acting like a professional and giving notice, but I wouldn't take it personally.
Reads like a classic project failure, with the classic failed project start: It was managed by someone who created "their own" database. i.e. A manager who thinks he knows better than the experts being hired, who overrides their estimates and recommendations, and who blows off any technical issues they raise because he "did it himself" in less time with an underpowered single-user tool.
I've worked on three similar projects -- only one succeeded. The one success was because the manager in question got yanked half way through and there was still enough time for a competent manager to be hired who'd let the team do their job properly.
I can think of a few other potentials, like the fabric of a chair sensing where the pressure points are and adjusting itself accordingly. I could see combining such a pressure-sense technology with a body-height cabinet for taking full-body measurements for custom clothing manufacture, and even more mundane uses like a more accurate "touch" scoring system for medieval recreationists.
For general robotics systems I can see more safety uses than practical -- wrap the limbs of the robot in this stuff so it can detect a collision that is out of it's field of view. (You just know that if we ever have household robots the kids are gonna climb on the things and one of the kids will end up hurt because the robot didn't see them.)
It looks like I'll have to do a bit of reading up on AI systems some day. As far as I knew, the issue was more one of neural learning networks vs. expert systems approaches.
Learning networks acquire and test connections and boundary conditions, keeping those which seem to be most relevant to the "world" as the AI learns. Expert systems are more predictable, but they only capture human knowledge for automation rather than learning new approaches.
From what you're saying, it sounds like the AI community breaks things down much farther. Personally I wouldn't have thought of constraint logicistics as being any different from tree pruning for an inference engine. Maybe algorithmically more efficient for certain classes of problems, but not functionally different.
Don't count on it. One of the famous government wastes in western Canada was a multi-year study on cow manure. Seriously. Not methane production, but manure. *shrug*
But I could imagine this maybe having impact on related fields that work with pressure-molded liquids or semi-solids. I'd think there are similar issues with producing an even coating/coloring for the glass part of a light bulb, for example. Sure it's not that they don't have ways of coating a bulb, but maybe his work might apply to a better way of embedding what is currently a coating, or applying it more evenly, or allowing the use of thinner glass, etc.
Not that I necessarily think it was worth 11 years, but having spent more than that on my own pet project I can well understand someone getting a tad fanatical about something no one else necessarily sees a use for. :)
The one I like is "Nobody will want to pull a Sony now."
Like it's a bad thing that corporations learn that invasive and damaging approaches to IP protection actually tick off consumers and cause lawsuits? The only surprise is that they had to learn by doing instead of just asking a few computer maintenance techs how they feel about the rootkit approach!
This was roughly 8 years ago. The town is north-west of Brampton/Bramalea.
It is no misinformation -- a number of techies I knew had planned on buying houses there until they found out the local cableco had been blocked from delivering high speed by the town's bylaws.
Most likely the bylaw has been fought and dumped since then, but the fact that it ever passed just goes to show the power of paranoid voters.
There is a suburb of Toronto that actually made it illegal for anyone to provide high-speed connections. IIRC, 64K is the limit because the "citizens" don't want their town to become a bedroom community of telecommuting techies.
Insane as it sounds, the town successfully passed the law. I don't know if it was ever overturned as an infringement of rights and freedoms, but the fact that a town could pass such an insane law means that a country could do the same.
Never underestimate the voting power of the retirees, blue collars, and low income people who hate the internet and even blame it (and computers in general) for the loss of jobs and the general dissolution of society.
Personally I find it amusing that management places so much emphasis on buzzwords like RAD, Scrum, and Extreme Programming. All most of these methodologies really do is emphasize what was known back in the sixties: a few skilled developers can code rings around a large team bogged down by communications and paperwork.
While both QT and GTK+ applications cross-compile to various *nix distributions and Windows, GTK+ does not require licensing fees to produce a closed-source application like Acrobat. While there is a "free" version of QT for Windows, it's only free for GPL applications IIRC. You can't even use it for development of a non-GPL application and defer the licensing fees until you have been paid by a client.
Another useful feature of Gnome is it's underlying reliance on CORBA technology. There are CORBA gateways for J2EE (and presumably .NET) that should allow you to build a native client against a TP-based J2EE server with Gnome. I don't believe you can do that with KDE/QT, at least not as tightly integrated/standardized as with Gnome. I don't know of any companies that have actually implemented such an application, but it would theoretically allow you to address the GUI performance issues while still maintaining a J2EE infrastructure that can also support web interfaces for external users.
Oops. Typo. That's Global Television, not CTV.
Ok, now this is getting bizarre. This week's episode of ReGenesis on CTV dealt with the old Spanish Flu virus being recovered from a corpse frozen in the permafrost of Nunavut. Either ReGenesis was dealing with very current information when it was scripted, the science teams are doing "No, it's really true!" reporting after the episodes, or someone isn't doing their fact checking.
http://www.regenesistv.com/
Google could just ensure that their test team is testing major vendor's hardware like Dell, HP, etc. After all, if you're talking about business and joe user functionality, you don't need to focus on 3D acceleration and such.
Google could just sink their cash into Novell/SuSE, RedHat, or Mandriva and provide a bundle that already works. Oh, wait, that's right -- you can already get Linux bundles with Java, OO/SO, etc.
So what's the "new" aspect you're suggesting, other than Google becoming involved in the marketing and distribution? What precisely is it that we need for a desktop GUI that isn't already in KDE and/or Gnome? 3D alpha-transparency spinners? Corona effects for the "glint" off metallic 3D lettering?
What Google could really provide in this area is some funding to improve the hardware support and configuration/maintenance utilities for components like configuring 3D support, adding/removing software, etc. I'm not talking about yet another front-end for RPM or APT, but some real improvement in reducing dependencies and manageability.
It's a way of implementing thin clients, but I don't know that it's a "crappy" way of doing so. The alternatives are facilities like X-terminals with an application server or equivalent services for Windows. As a browser-based thin client at least offloads a lot of the display updates to client-only traffic instead of network traffic, it would make a rather substantial difference in the bandwidth required by a thin client to edit documents.
Besides, don't confuse "potentially useful" with "good idea." It's just an interesting approach, and it seems to work ok for very simple documents like email text and web forum editors.
Linus is talking about hardware specs, then comparing them to mixed model specs like OSI. He further generalizes that specs are useful for talking about implementations, but not for actually doing them.
The problem is that he's ignoring how important that discussion framework is when trying to bring new people into a project, or trying to maintain components after the lead engineer(s) are gone. A good spec provides an invaluable model of how to think about the implementation, not a written-in-stone detail of how to write the implementation.
But he is right about bad specs. I once worked with some old Sperry mil-spec hardware (AN-UYK502.) The compilers produced code based on the hardware spec, but the real hardware implementation had reversed the registers for the results of division. Every piece of code produced by the compiler had to be manually tweaked to correct the mistake, but because the compiler met the spec, the vendor wasn't allowed to fix the problem. To the military mindset, it was the hardware that didn't meet spec and needed to be fixed.
Reliability is another issue. If I have the documents on my local drive, I can always retrieve them from backups if the system goes down. But if the internet link is down, it doesn't matter that I've got other computers available -- I still can't get to my documents.
Where I do see such technology having potential use is for the intranet, where it's already common for documents to be stored on a central server instead of local hard drives.
Going offshore isn't much protection if the US gets their way:
http://news.zdnet.co.uk/business/legal/0,39020651, 39220179,00.htm
Microsoft, *IAA, etc. know they can't win against offshore firms and open source under the current global legal system. They're pushing hard to have US laws (and presumably the US patent/copyright databases) applied globally.
The manager claims to have run SAP on AIX for several years, and to have been comfortable with that. Given the minimal pricing difference between Linux/x86 and AIX/POWER for IBM bundles, why wouldn't they have gone with AIX in the first place?
Don't get me wrong -- I've found Linux to be very solid for a number of years. But when you're dealing with package compatabilities and interdependencies, sometimes a commercial OS is easier to track and maintain.
It's also not very clear whether the core problem is with SAP or RedHat. Given the engineering squads that are required to customize an SAP installation, I'd really hesitate to presume the problems couldn't have been faulty customization rather than either of the core products.
The only so-called "database" that emphasizes it's GUI is Access. Every other vendor/product I'm aware of relies of separation of duties and doesn't try to roll user interfacing into what is rightly a back-end service.
Administration tools for commercial and OSS databases may be easy for small sites and novice DBA's that don't know their tools, but large applications rely on database scripts to handle configuration, not GUIs. The reason is simple: you can't put a mouse click into CVS/RCS/SCCS/???.
All laptops, workstations, and PCs provided by the company are the property of the company. They are provided to help you do your job.
IIIR has no website which can be located by Google, Yahoo, or any other search engine that I've used.
All that comes up are some investor reports like OneSource, which reports them as having a whole seven employees.
Trademarks are not automatically international, and the mere presence of the "G" before "Mail" is not a trademark. Trademarks are either specific spellings (with/without hyphens), logos, icons, color combinations, etc.
The lawsuit sounds like basic "trolling for dollars", legal style.