Did anybody notice that the United States National Security Agency is buying encryption software from a Canadian company? Is this the same United States that refused to allow products using good encryption to be exported because they were considered military weapons?
I am not flaming Canada; I work with several Canadians and they are all nice and knowledgable people. I just noticed the inconsistencies in our policies.
Disclaimer: I am a citizen of the USA, and I hope that this trend continues. I would really like all our government agencies to use the best global software, not just our homegrown insecure proprietary systems.
Bill wondered if Phoenix, the BIOS company, still existed. Can any regular Slashdot reader have missed all the recent articles about the Mozilla variant named Phoenix being told by Phoenix the company to stop using their trademark?
Not only does Bill not read Slashdot, it is doubtful he has any clue what is happening in the computer world. That is why MS is always playing catch-up.
I absolutely, positively, need for each single cell in my body to have its own IP address.
Depending on your size and who did the calculations, there are between a trillion and a quadrillion cells in your body. A quadrillion is around 2^50. We could assign an IP Address for every human on the planet (approx 2^34). With IPv6, there are still 44 bits unused.
I highly recommend keeping your cells' addresses in their own domain. If you do need to connect to other resources, use a NAT to keep the addresses private. You also want a firewall, such as skin, and maybe some input filters, such as teeth and gastric acid. These have probably already been installed, but you do need to maintain them. And be careful with your choice of OS: you do not want a virus invading from the internet, and a crash could be fatal.
Parent posted to World's 10 Tallest Buildings which shows pictures of all 10 buildings (including Taipei 101) on a ruled page that shows the height in meters.
The graphic is very pretty. It also shows that only the Sears tower has a full floor over 425m. And that if the permanent antennas were redefined as spires, it would still be the tallest.
US are #1! I mean, we are #1. I mean, well, I am not an architect or associated with the construction industry in any way, but national pride says we are still the highest (in every sense of the word.)
Taipei 101 will hold 3 of the World's Tallest Building titles when it is topped out: Tallest to structural top, Tallest to roof and Highest occupied floor.
Taipei 101 now holds the title of the world's tallest building measured to the roof, replacing the Sears Tower.
The articles do not give a number for Highest occupied floor, but: 1667 - 197 (spire) = 1470 feet. The Sears Tower is occupied to 1431 feet.
"There are only two investors in this deal: BayStar Capital and the Royal Bank of Canada.
"I think people will try and come to the conclusion that Microsoft is somehow involved in this deal, but I can tell you with great certainty that Microsoft was not involved with this investment,"
Let us assume he is telling the truth (because we know nobody would lie to a reporter just because MS suggests it and gives them money. Where's a link to the Switch campaign?)
So MS controls the Royal Bank of Canada? Maybe not.
About Longhorn, "This release is going to be driven by technology, not by a release date. Which probably means it is going to be late." If there will not be a release date, then how can it be late?
"We have a lot more understanding of database technology these days" That should scare anybody who is using MSSQLServer.
From the article: One thing that seems to slow down the next release of Windows is the much talked about data storage system WinFS, technology designed to make information easier to find and view. Since it is based on the next version of SQL Server or Yukon, the system will essentially function as a relational database. Bill: "We will have pointers in the data like a URL or weblink. URLs are a perfect tool for this, but in previous databases we really had a problem with them. They screwed up the query semantics." So every flaw in MSSQLServer will affect the file system. And will be accessible using URLs.
"How could we ignore the browser? The Explorer is fully integrated with the operating system, take it away and the OS grinds to a halt." Don't you love that every flaw in the browser affects the whole OS? (On my system, MSIE is the only application that grinds the OS to a halt.)
This quote was responding to the lack of feature updates for MS Internet Explorer. Is this from the same company that announced there will never be a new version of Internet Explorer?
"To be honest, I haven't heard from Phoenix Technologies for over five years. Are they still in business?" Bill does not follow technology news? Slashdot posted many articles about the Phoenix trademark issue when there was a Mozilla variation using the name. This answers whether Bill reads Slashdot.
"But apart from Photoshop, I can't think of desktop applications where you would need more than 4 gigabytes of physical memory" But won't 4 GBs of RAM be required just to load Longhorn and the then-current MSOffice? MS seems to be one of the main drivers of the need to upgrade consumer hardware. We know this quote is going to haunt Bill for a very long time.
"Critical security patches should be applied with the speed of the internet." Viruses are already applied at the "speed of the internet." Patches need to be even faster.
"We used to send megabytes of software to fix a 20 byte file" But the viruses were already small!
"We invented personal computing." Remember Apple? Atari? Commodore? Tandy? And anybody else that sold a personal computer before 1981? He could say that he brought personal computing to the masses, and taught them the definition and joy of "reboot".
--- I skipped how he says MS is releasing patches faster than the Linux community. MS might be releasing MORE patches, but faster? Read the article if you want a laugh.
If you are going to troll, at least do it correctly.
--- The problem with finding Domino Administrator jobs is: 1. The people in those jobs are rather highly paid for an computer administration position. In 2000, Certified Lotus Professional System Administrators averaged $89,000. 2. They do not need to worry about viruses beyond choosing and installing a mail filter/virus protection, since no viruses have hurt Lotus Notes yet. The virus protection checks those virus-prone Word files, and helps if users are using MSOutlook as the mail client. 3. The number of administrators needed for a company running Notes and Domino is much less than the same company running Exchange. This is anecdotal from personal experience. I know a 500 person company that grew from one person doing Notes Admin work part-time in a computer department of 2 people to 2 full-time Exchange Admins with 10 people in the computer department, at a time when the company was shrinking. A 30,000+ employees company went from 10 Notes administrators maintaining their own servers to 60 Exchange administrators with the servers maintained by a different group. This is only the Administration side, application development costs skyrocketed while application rollouts almost disappeared after the switch.
Domino Administrators are happy, and companies of any size do not need very many of them. There is little turnover, and so there are few jobs to be filled. (Besides, who is going to quit with today's job market?)
--- To be on-topic, Lotus/IBM releases updates at least quarterly. The updates usually add functionality, and fix crashes due to very unique circumstances. I only remember 2 that were for security issues. One was only an issue if the option to use MSIE as the browser was selected. The other was only an issue if Notes Designer was run in a certain configuration without a firewall. None of the updates are "critical". I just upgraded one large company's server from Domino 5.0.2 because the hardware was being replaced.
To be fair, while Domino is a platform, it is not an OS. It relies on Unix, Linux, or MSWindows for its file protection. If you are running MSWindows, you may need some of these patches. Then again, if Domino is only running mail, web applications, and Notes client applications, you can turn off most of the vulnerable MS services.
The Lindows example is a definite "distribution" sold to many people who care about the GPL, the NDA was probably click-wrap, and Lindows needs the goodwill of the open source community.
The GPL states: "You may not impose any further restrictions on the recipients' exercise of the rights granted herein."
If IBM sold BigCompany a CD containing some software for the cost of the media ($1,000,000) and BigCompany promised not to request the source or redistribute the software, there is no one to complain about it. What could happen? Bruce calls IBM and says "You didn't give BigCompany the source." IBM says "They didn't want it." Richard calls BigCompany and says "You need the source." BigCompany says "We have outsourced every IT position to IBM. They have the source. We don't want it. And if we had it, we wouldn't give it to you anyway." No blood, no foul.
--- Hey, we are back to the point of my original post about SCO revoking IBM's Unix license. The hypothetical situation was that IBM needed to replace AIX ASAP. IBM does not care about software. They do care about competitive advantage, and they hate SUN. They are willing to kill the best software available today just to attack SUN.
Would they release their "proprietary and different" technology under GPL? Maybe.
They could start their new OS with a BSD. Then they could keep it proprietary.
But releasing it under GPL could be another attack on SUN. If SUN wants the changes, then SUN must recreate Solaris under GPL. Then SUN loses the advantage of their "proprietary and different" software. The end result is that there is no more proprietary Unix, and SCO goes down the drain.
MSWindows2000 was released on February 17, 2000. RedHat 9.0 was released on March 31, 2003 (All release dates for RedHat are from this link.)
So Redhat had over 3 years to fix holes in the distribution, while crackers had 3 more years to find holes in MSWindows2000. If you want to play fair, - use Windows2003 released April 24, 2003 and RedHat9.0 (24 days between the releases), or - use RedHat 6.2 (released March 8, 2000) (19 days difference from MSWindows2000) or maybe RedHat 7.0 (released August 28, 2000).
If you want to compare RedHat9.0 and MSWindows2000 today, you should fully patch both of them.
I believe RedHat will still win any of the fair tests, but if you use unpatched RH6.2, it will get cracked. It will just take longer because there are tons of script kiddies just sending easy URL cracks at every web server hoping it is running MSWindows.
My mistake, bad example: I forgot the Apache License is BSD-style.
Your mistake: bad example: MSWindows is not GPL. Their licenses are extremely draconian. Anybody releasing their source code would have very serious legal difficulties. But as a consultant, companies often share their proprietary licenses with me so I can work on their systems. As long as someone pays for the licenses, the proprietary software companies do not care whether the license is used by an actual employee or a partner. NDAs do not affect this practice, just financial wisdom.
You are correct: Patent law does not apply here. (For the record, our software is completely home-grown, and does not include code from any other project, GPL'd or not.)
The companies in question engaging in the NDA covered distribution changed the NDA's and published source after chats with FSF's legal counsel, even if they were just temporarily engaging in the practice for understandable reasons.
Do you have links? I do not remember this method being used. I remember several cases where GPL'd software was distributed, and the FSF was able to get the source distributed too. And you state that the NDAs needed to be changed, so maybe they were legal, but did not keep the spirit of the GPL? Most companies want to keep on the good side of the open source community, and will bend to keep them happy. The good publicity is worth more than any advantage from keeping the code secret for a few months.
This thread was speculating about a possible reason (losing the Unix license) that IBM would consciously break the spirit of the GPL. If IBM really needed an Unix-like OS in a hurry, they would probably start with a mature BSD rather than the weaker and GPL-encumbered Linux. (That is what Apple did.)
As I said, IANAL. I am interested in the links that this has been tried. But I seriously doubt IBM would bother trying it.
There are several good reasons for MS to delay Longhorn. (Please remember that I believe MS will enter its death spiral within 2 years.)
The first reason is financial tactics. MS has managed to get companies to pay for 3-year licenses. They have already paid, so why give them anything? MS will release the next OS at the moment just after most of these licenses need to renewed. "Renew now and the next OS is included." generates much more income than "You just received everything we expect to release for the next couple of years. Would you like to renew?"
The second reason is publicity. Talking about Longhorn puts MS in the news, which always helps the stock price. As long as it is vaporware, they can do a press release anytime there is not a new virus threat in the news. Releasing the software slows the PR.
The third reason is their business operations. They are reducing the number of programmers in the U.S. and hiring programmers in India to do the work. The Indian programmers probably keep saying things like: "This code you sent us does not make sense. There is no way it could ever run, and if it did, it would cause memory leaks and crash anyway." Moving code production overseas always incurs some delays, but it will take a long time for anybody outside Redmond's groupthink to understand this code.
The final reason is that Bill knows his company is declining. Why spend the money to develop software when your company will not exist to release it?
I always thought Pascal was another MS-related casualty.
In the 80s, everybody was talking about Pascal being the next great language. There were already descendants like Modula2 and ADA that were gaining market share. (This upset me because I learned Pascal, then C, and C looked more logical, so I hoped it would win.)
Then MS introduced QuickBasic, which turned into VisualBasic. (I have a book published by IBM stating "VisualBasic is a trademark of IBM.") Around 1992, you could probably translate programs from Pascal to VB or v.v. by substituting the keywords. The structure and the syntax was almost identical.
VB became the language of choice for beginners. It filled the same ecological niche as Pascal, so Pascal withered.
--- The difference between Pascal-based languages and C-based languages:
Pascal-based languages allow "procedures". Procedures can define multiple inputs and multiple return values. VB and the Windows API follow this syntax. It is easier for beginners to understand, but maintenance is difficult because the programmer needs to remember which variables can be changed safely and which will be returned.
C-based languages insist everything is a function. Functions have 0 or 1 return value.
When the C algorithm for returning multiple values by passing a pointer to a struct became difficult for maintenance, they invented classes, which allowed the multiple values to be retrieved. That was C++, but most early tutorials explained it as an addition to C, rather than a move to OOP. This is still evident in the ease of mixing C and C++ code in the same program.
Then Java added easier memory management and notification of errors (and reintroduced the virtual machine for sandboxing.) It follows the C++ model, but prefers OOP design.
SCO has killed Unix-branded Unix. Can you imagine any company willing to buy a license after SCO started suing some of the licensees (IBM, SGI)? The effort to start a new OS from scratch is enormous. (See BeOS.) Any new OS is going to be based on BSD or Linux. (See Apple.)
There will never be a new SystemV-based Unix, which means SCO can never sell another license. And most of the current licensees (not including MS and SUN) paid AT&T in full for perpetual licenses, so there is little revenue from the current licensees.
Lawsuits as a business model is all that SCO has left.
limiting the receivers' rights is not acceptable under the GPL. Section 6: "You may not impose any further restrictions on the recipients' exercise of the rights granted herein.". This applies to any kind of limitations, ranging from patents to NDA's.
There is no spoon. (Sorry.) There is no receiver. There is no recipient. The nice part (from a legal perspective) is that an NDA can make a company a development partner which is part of the company for purposes such as patents.
So my company rewrites half of Apache. We do not want to distribute the code, but you are a friend, so we will lend you a server with everything configured. You ask about the source code, and we present the NDA stating that you are a development partner, will be testing the software in a production environment, and have no rights to any of the modifications.
Under U.S. patent law, that is legal, and there is no distribution of software because those companies are partners. So there is no "recipient" to have rights restricted. So there is still a hole in the GPL.
--- I like the GPL. I think it is as good as it gets for sharing software.
And this hole cannot be closed. Try adding something stating the moment you change something you are required to tell the world. Who would touch software if you had to issue a press release every time you changed a line of code? You must be able to work on it without being required to distribute.
(IANAL. Is this a good place to put that?)
--- Why I know this We are developing software. We think we can get patents. We cannot discuss the features that might be patented with anybody with whom we do not have an NDA without losing our ability to patent those features. Well, we would have one year to file in the U.S., but we would immediately lose the right to file in Europe. The NDAs protect our rights to file, and we insist all potential customers sign them before we give a sales presentation. This limits to whom we can sell, so we want to get the patents filed so we can skip this step. It is possible for a company with the clout of IBM to force every "customer" to sign an NDA to become a "partner" to avoid the "distribute" clause.
Again, I am not saying this would ever happen, just that it is possible.
Have you even read the GPL? Any modifications to the Linux kernel are necessarily GPL'd as well.
Have you even read the GPL? Did you actually read my post to which you are responding? The GPL has a very narrow scope that is easily avoided.
The paragraph in my post just before the one you criticized explains the loophole in the GPL. Yes, modifications to GPL'd software are GPL'd. But the GPL only applies when you distribute software. If you make changes to Apache, and only use it at your company, the software has not been "distributed" and you are not required to provide the source to anybody. If you give or sell it to another company, you must make the source available to that company, but you could contract that they not distribute the software or the source to anybody else. An NDA (Non-Disclosure Agreement) allows you to share technology without allowing that technology to be distributed to others. A well-written NDA could be used to sell GPL'd code without providing source code.
Most software comes with contracts/licenses that state that it may be used on a certain number of machines for a given time period and may not be shared. All the DRM news is about companies trying to enforce those contracts through technology, even though the technology does not allow for fair-use.
Even the SCO circus is about the same battle. SCO thinks they own the rights to Unix, and they believe that the licenses to sell Unix implies that the licensees must protect Unix. Given that, then any code written to improve Unix cannot be shared with OSes that compete with Unix. Yes, they have warped minds, but it does make sense if you take the correct drugs.
I was writing about a hypothetical future where IBM dumps all of their technology that makes AIX special into a Linux-based OS. It is likely that much of it can be handled by proprietary programs outside the kernel, but anything related to making the kernel robust on 64+ processors would require kernel modifications. Nothing I said implies that IBM has not obeyed the spirit of the GPL so far, but in this hypothetical future situation, they would be giving away much of their private technology.
It is possible for IBM to take Linux and turn it into the next version of AIX. My point was that if they did this, they would want to protect their technology, and there is a way for them to do it. I was not implying that IBM would violate the spirit of the GPL. I was not implying that these events would ever happen, since that requires IBM's Unix license to be revoked and I doubt that even our courts would allow that. IBM sells hardware; the OS is just an extra, so I doubt they would even care, except for the existence of HP and SGI and the other Unix sellers who would not mind reading the source to an OS that included all of IBM's tricks.
Take this as completely theoretical. I believe SCO is evil and IBM is going to destroy them. But...
What if SCO is able to revoke IBM's license to sell Unix? The contract is broken. IBM pays $3B. And IBM needs a quick replacement to convert all those AIX installations.
The contract is broken, so IBM no longer has any obligation to protect Unix. And the customers demand all IBM proprietary technology must be in the replacement that the courts have ordered IBM to provide.
IBM cannot use Unix. They need an OS that can replace AIX, meaning it can run their customers' Unix applications. Can this be done with OS/390? OS/400? OS/2? Or will IBM immediately pump all that technology into Linux?
IBM Linux could become suitable for big iron in a very short time.
Not that all that technology must be added to RedHat or SuSE. The big loophole in the GPL is that it only matters if you distribute the binaries, and that "distribute" means customers, and U.S. law uses NDAs to change a customer into a development partner that is legally associated with the developer so the GPL is not relevant. But it would only take one customer to demand a replacement without signing an NDA to force IBM to GPL all this code. Will any company chance their information systems becoming obsolete because they did not sign a contract that does not affect themselves while IBM is offering a big discount?
Let us assume that IBM plays nice and GPLs all changes to the Linux kernel. They could still keep the code that uses those changes proprietary, but the kernel is improved. (And IBM offers Torvalds $1B to oversee the project, and he would take it because it would mean fantastic progress for GPL'd Linux.)
I have no idea how this situation helps SCO. I think it would be the final straw and "Unix is dead" would appear everywhere. I doubt SCO has put any thought into the consequences of their actions, other than they might get $3,000,000,000. I know that much money would make me happy, so maybe that is the only point, and they expect Unix to die anyway.
Yes, the Notes client has always been graphical, even in the 80s. The screen shots you linked do not express the power of Notes in a corporate environment. Today I prefer it for rapid development for web applications, although the client is still useful. Notes has the ability to consistently deploy secure and distributed applications with less than one week from idea to use.
I am designing a new platform, but I am starting by not trying to build a direct replacement for Notes. The new platform will have a superset of the features of Notes. We can write converters that can move Notes applications to the new system, rather than build something that uses the Notes file formats directly.
I sometimes hope to find an open source project that can be used as the base for the new system. From the text list of functions, Notes does everything Citadel does, so Citadel would have filled many of the functions needed for the replacement. I accept that the architecture is too different.
To the ACs who noticed that they should have been &&s, not ||s, you are correct.
About the lousy comments, you are also correct. Some of my code does look like this, because I would start by writing an English version, comment it out, and then insert code. The redundant comments get removed as time passes and the code is revised. My comments tend to focus on the business reasons, with a few to explain more complicated code or to warn about possible failures. I agree that "Code Complete" is a good book for programmers who want to move to category 4.
I wrote the example quickly to point out that using chained "||"s in an "if" were not necessarily bad, but if it made the code difficult to understand, then it was bad. So there was no fifth category of programmer, since someone who coded this way would still fit into the first 4 categories. (A case could be made that many who program this way would be category 1.)
About the lousy code, sorry. My first example had functions with names like "action1()", and could have done anything. There were 4 examples of code, with the last one attempting to get cute by finding a possible business use for the code. I redid the other examples to match the last version, then deleted the middle ones to save space. I had not noticed that I needed to switch from ||s to &&s to make sense. I almost never code that way, since it does make the code less understandable. Try working on code you wrote 10 years ago. You will wish it was very easy to understand. Then imagine someone else trying to work on it. I do not want the negative karma from having that other programmer curse at me.
Your fifth type of programmer fits in either the third or fourth category.
3. Those who write good code most of the time, but who don't fully realize their limitations If you write code that is difficult to maintain for the fun of having written it, you belong in the category of those who should be limited. Let another programmer review your code before it is committed. If any line takes more than a minute to understand, he either rewrites it or passes it back.
4. Those who really understand the language, the machine architecture, software engineering, and the application area, and who can write textbook code on a regular basis. Part of software engineering is knowing that code will require maintenance someday. The code is easy to understand. The code itself code be complicated, but comments make certain the next programmer knows exactly what is happening.
severely abuse short-circuit evaluation It is the difference between: for(i=1;((x=getEmployeeNumber(i))>0)||((y= getDaysSinceRaise(x))>0)||((z=calculateRaise(y))>0 ));i++) notifyManager(x,z);
And
//Give raises until someone does not deserve one or an error occurs for(record=1; ((empnum=getEmployeeNumber( record))>0)//get next employee ||((days=getDaysSinceRaise(empnum))>0)//get number of days, errors return 0 ||((amount=calculateRaise(empnum, days))>0));//get amount of raise, verify it is positive number record++)//Next record notifyManager(empnum, amount);//tell the managers so they can verify before telling the employee.
This is the same code, but the latter is more understandable. It is not the code that matters, it is how you use it.
I had never heard of Citadel. The function list sounds like Lotus Notes without rapid application development, or any development environment for that matter, although the homepage says Citadel can be used as a datastore for applications. And Citadel has native ability for accessing file systems. Thanks for the link. I may install and review it.
Open Source replacements for Lotus Notes try to recreate all the functionality while keeping all the flaws, similar to Linux windows managers trying to become MSWindows, although at least the WMs include improvements like multiple desktops. Citadel is open source and includes most of the functionality of Lotus Notes. If someone created a development environment that followed the patterns used in LN, it might serve as a complete replacement. Of course, I have not installed Citadel yet, so it is possible that the architecture makes it completely inappropriate for this.
Do you have any experience with the Citadel developers? They ask for "small patches". Any chance they would add a major enhancement, or would it be better as a separate project?
--- About the Flamebait: I probably should have put the first line in parantheses. I thought the parent post was quoting somebody and forgot to turn off the italics. Since it is standard to put quotes from other posts in italics, I thought a gentle reminder would be OK. I was not posting just to complain; I really expected a +1 Funny if it was modded at all.
I changed my sig because my 120th post was the first to be modded down, and I felt it was incredibly funny that the post was modded "flamebait" while talking about sun power. OTOH, I have been getting moderator points 2 or 3 times per week, and I have not received any mod points since that post. Or it could just be that the Slashdot system has been changed, since the "Quote of the Hour" has not changed in 2 days.
If I had a TV, I would want to tell my television to get the current streamed content from: tv://hbo1.hbo.com The television has my logon, and uses it to access the content. HBO gets my IPAddress and logon, and can be certain that I am only receiving the content at one location at a time. They could track exactly what I watch, but that is a different issue. I could pay HBO by the hour, or a monthly fee.
The cable, telephone, and ISP companies should not be involved in the transaction. Why should these infrastructure companies have anything to do with my content?
Because they own the wires? I do not want a monopoly to have any control over my content. Yes, they made the investment. Yes, they should make their money back, charge maintenance costs, and profit for as long as wires are still useful. But let me see what I want to see, not what they decide to show.
I am using HBO as my example because they have shows that I would watch, if I could get just those shows. HBO could charge up to $1/hour or $20 per month, and let the first episode of each new series be free. I saw "The Sopranos" and "Six Feet Under" a few years ago. I would pay $8 per month to receive those 2 shows for a few months of each year (if I could watch them when I want. I still would not plan to be home at a certain time just for entertainment.) But I will not pay for tons of content that I do not want tempting me.
If this happened, there would probably be tv://sopranos.hbo.com/season3/episode5.show i n the near future.
I doubt anybody in the industry wants this to happen. New shows would be much more difficult to push. And shows with loyal followings like ST:TOS would be harder to cancel when exact numbers of watchers is known, and revenue can be attributed directly to each watcher. Shows with limited audiences could survive by basing their production budget on the revenues. Amature production companies would have a chance.
The cable companies would still exist as both the infrastructure (probably killing the telcos) and financial centers for collecting subscriptions, but the public would have more control.
Why shouldn't infrastructure companies make money? They made the investment; they should get the returns.
The cable companies could still negotiate with the content providers for discounts and bundle everything and act as a clearing house for subcriptions. Unless they decide they do not want to carry the WB or that amature station. Then the "subscribers" could buy the programs directly from the producers. Most people would prefer to have a single bill for most of their television channels, and the discount for bundling the usual channels would be a considerable savings. But this way the subscribers have choices.
Thank you. I never took the time to see if anybody else was using the format. I used it. People understood it. It was better than what I was taught. I trained myself to always use it.
Around 1996 I was marking computer file names (specifically my resume and backups of programs in development) with the date. I switched to year first so they would sort correctly. I was working on international projects. Once I realized other people had no problems interpreting them, I started using them everywhere. In early 1999, I started using a 4-digit year to avoid the Y2K issue, and I have not written a digit-only date in any other format since then. (My resume's cover letter does contain the "Month D, YYYY" at the top.)
Using this format does generate converstaion. People usually start with "That's different, but logical. Does everybody understand it?" Now I can say it is a standard and point to ISO 8601. Thank you.
--- To the grandparent post, I am in the US, but I started using this format because I was working internationally. People in the US assume MM-DD-YYYY, but several of my programs are used globally, and the use of this format saves much confusion. I convinced management that this format would be good when we had a major issue about a deadline of 4-6-2000 involving some Europeans.
Separation of hardware and software and content. (like the separation of church and state.)
Infrastructure companies made the investment to install the wires to an area. Let them make their money from renting those wires to other companies that provide services.
The "Cable" companies are winning because they laid the wires for the sole purpose of providing content, and were regulated as content providers. Now they are branching out, but want to keep the laws as they were when they were a content provider that had to create its own infrastructure. They should be considered an infrastructure company that provides content, or better, be forced to separate the functions into distinct financial entities.
The power companies went through the same issues in the last decade. In my area (PA), you can buy your power from several companies, and PECO is responsible for its delivery. The delivery charges are more than the power generation charges, but at least you have some choice.
Let the infrastructure company charge standard rates to all who want to use them, including: - the sister company that sells service, and - the sister company that sells content. - any company that wants to use the wires and can afford to pay the standard rates.
I have 2 choices for information wires to my house: Verizon or Comcast. Neither can keep a connection alive for a week. Verizon has been forced to allow others to sell DSL service over their lines. Comcast has not. I prefer cable, which means I have only one choice for my ISP. It would be better if each functon was separate: - one company for the physical connection. - choice of companies for the software connection. - many choices for service (mail, website hosting) and content providers (news, search sites, CableTV)
HBO could run its own service. You could subscribe to HBO, SF, and UPN without getting QVC and CBS. (I do not watch any television, so I may not have the correct channels for cool vs. uncool.)
Let the government regulate and monitor any company that owns the wires to make certain they are priced appropriately.
Slashdot Mirror
Did anybody notice that the United States National Security Agency is buying encryption software from a Canadian company? Is this the same United States that refused to allow products using good encryption to be exported because they were considered military weapons?
I am not flaming Canada; I work with several Canadians and they are all nice and knowledgable people. I just noticed the inconsistencies in our policies.
Disclaimer: I am a citizen of the USA, and I hope that this trend continues. I would really like all our government agencies to use the best global software, not just our homegrown insecure proprietary systems.
This was answered in my post about an interview with Bill.
Bill wondered if Phoenix, the BIOS company, still existed. Can any regular Slashdot reader have missed all the recent articles about the Mozilla variant named Phoenix being told by Phoenix the company to stop using their trademark?
Not only does Bill not read Slashdot, it is doubtful he has any clue what is happening in the computer world. That is why MS is always playing catch-up.
I absolutely, positively, need for each single cell in my body to have its own IP address.
Depending on your size and who did the calculations, there are between a trillion and a quadrillion cells in your body. A quadrillion is around 2^50. We could assign an IP Address for every human on the planet (approx 2^34). With IPv6, there are still 44 bits unused.
I highly recommend keeping your cells' addresses in their own domain. If you do need to connect to other resources, use a NAT to keep the addresses private. You also want a firewall, such as skin, and maybe some input filters, such as teeth and gastric acid. These have probably already been installed, but you do need to maintain them. And be careful with your choice of OS: you do not want a virus invading from the internet, and a crash could be fatal.
Parent posted to World's 10 Tallest Buildings which shows pictures of all 10 buildings (including Taipei 101) on a ruled page that shows the height in meters.
The graphic is very pretty. It also shows that only the Sears tower has a full floor over 425m. And that if the permanent antennas were redefined as spires, it would still be the tallest.
US are #1! I mean, we are #1. I mean, well, I am not an architect or associated with the construction industry in any way, but national pride says we are still the highest (in every sense of the word.)
Taipei 101 will hold 3 of the World's Tallest Building titles when it is topped out: Tallest to structural top, Tallest to roof and Highest occupied floor.
Taipei 101 now holds the title of the world's tallest building measured to the roof, replacing the Sears Tower.
The articles do not give a number for Highest occupied floor, but:
1667 - 197 (spire) = 1470 feet.
The Sears Tower is occupied to 1431 feet.
"There are only two investors in this deal: BayStar Capital and the Royal Bank of Canada.
"I think people will try and come to the conclusion that Microsoft is somehow involved in this deal, but I can tell you with great certainty that Microsoft was not involved with this investment,"
Let us assume he is telling the truth (because we know nobody would lie to a reporter just because MS suggests it and gives them money. Where's a link to the Switch campaign?)
So MS controls the Royal Bank of Canada?
Maybe not.
Douglas T. Elix, Senior VP of IBM Global Services, is on the Board of Director of the Royal Bank of Canada. (Actual list is available in a PDF from that page.)
New conspiracy theories!
IBM is funding SCO.
But why?
Bill's quotes are in quotes.
About Longhorn, "This release is going to be driven by technology, not by a release date. Which probably means it is going to be late."
If there will not be a release date, then how can it be late?
"We have a lot more understanding of database technology these days"
That should scare anybody who is using MSSQLServer.
From the article: One thing that seems to slow down the next release of Windows is the much talked about data storage system WinFS, technology designed to make information easier to find and view. Since it is based on the next version of SQL Server or Yukon, the system will essentially function as a relational database.
Bill: "We will have pointers in the data like a URL or weblink. URLs are a perfect tool for this, but in previous databases we really had a problem with them. They screwed up the query semantics."
So every flaw in MSSQLServer will affect the file system. And will be accessible using URLs.
"How could we ignore the browser? The Explorer is fully integrated with the operating system, take it away and the OS grinds to a halt."
Don't you love that every flaw in the browser affects the whole OS? (On my system, MSIE is the only application that grinds the OS to a halt.)
This quote was responding to the lack of feature updates for MS Internet Explorer. Is this from the same company that announced there will never be a new version of Internet Explorer?
"To be honest, I haven't heard from Phoenix Technologies for over five years. Are they still in business?"
Bill does not follow technology news? Slashdot posted many articles about the Phoenix trademark issue when there was a Mozilla variation using the name. This answers whether Bill reads Slashdot.
"But apart from Photoshop, I can't think of desktop applications where you would need more than 4 gigabytes of physical memory"
But won't 4 GBs of RAM be required just to load Longhorn and the then-current MSOffice? MS seems to be one of the main drivers of the need to upgrade consumer hardware. We know this quote is going to haunt Bill for a very long time.
"Critical security patches should be applied with the speed of the internet."
Viruses are already applied at the "speed of the internet." Patches need to be even faster.
"We used to send megabytes of software to fix a 20 byte file"
But the viruses were already small!
"We invented personal computing."
Remember Apple? Atari? Commodore? Tandy? And anybody else that sold a personal computer before 1981?
He could say that he brought personal computing to the masses, and taught them the definition and joy of "reboot".
---
I skipped how he says MS is releasing patches faster than the Linux community. MS might be releasing MORE patches, but faster? Read the article if you want a laugh.
Your search - Domino Administrator wanted - did not find any openings.
--Monster.com
Monster.com has 26 jobs listed.
Dice.com has 20 jobs listed.
JustNotesJobs.com has 16 jobs in the U.S.
If you are going to troll, at least do it correctly.
---
The problem with finding Domino Administrator jobs is:
1. The people in those jobs are rather highly paid for an computer administration position. In 2000, Certified Lotus Professional System Administrators averaged $89,000.
2. They do not need to worry about viruses beyond choosing and installing a mail filter/virus protection, since no viruses have hurt Lotus Notes yet. The virus protection checks those virus-prone Word files, and helps if users are using MSOutlook as the mail client.
3. The number of administrators needed for a company running Notes and Domino is much less than the same company running Exchange. This is anecdotal from personal experience. I know a 500 person company that grew from one person doing Notes Admin work part-time in a computer department of 2 people to 2 full-time Exchange Admins with 10 people in the computer department, at a time when the company was shrinking. A 30,000+ employees company went from 10 Notes administrators maintaining their own servers to 60 Exchange administrators with the servers maintained by a different group. This is only the Administration side, application development costs skyrocketed while application rollouts almost disappeared after the switch.
Domino Administrators are happy, and companies of any size do not need very many of them. There is little turnover, and so there are few jobs to be filled. (Besides, who is going to quit with today's job market?)
---
To be on-topic, Lotus/IBM releases updates at least quarterly. The updates usually add functionality, and fix crashes due to very unique circumstances. I only remember 2 that were for security issues. One was only an issue if the option to use MSIE as the browser was selected. The other was only an issue if Notes Designer was run in a certain configuration without a firewall. None of the updates are "critical". I just upgraded one large company's server from Domino 5.0.2 because the hardware was being replaced.
To be fair, while Domino is a platform, it is not an OS. It relies on Unix, Linux, or MSWindows for its file protection. If you are running MSWindows, you may need some of these patches. Then again, if Domino is only running mail, web applications, and Notes client applications, you can turn off most of the vulnerable MS services.
The Lindows example is a definite "distribution" sold to many people who care about the GPL, the NDA was probably click-wrap, and Lindows needs the goodwill of the open source community.
The GPL states:
"You may not impose any further restrictions on the recipients' exercise of the rights granted herein."
If IBM sold BigCompany a CD containing some software for the cost of the media ($1,000,000) and BigCompany promised not to request the source or redistribute the software, there is no one to complain about it. What could happen? Bruce calls IBM and says "You didn't give BigCompany the source." IBM says "They didn't want it." Richard calls BigCompany and says "You need the source." BigCompany says "We have outsourced every IT position to IBM. They have the source. We don't want it. And if we had it, we wouldn't give it to you anyway." No blood, no foul.
---
Hey, we are back to the point of my original post about SCO revoking IBM's Unix license. The hypothetical situation was that IBM needed to replace AIX ASAP. IBM does not care about software. They do care about competitive advantage, and they hate SUN. They are willing to kill the best software available today just to attack SUN.
Would they release their "proprietary and different" technology under GPL? Maybe.
They could start their new OS with a BSD. Then they could keep it proprietary.
But releasing it under GPL could be another attack on SUN. If SUN wants the changes, then SUN must recreate Solaris under GPL. Then SUN loses the advantage of their "proprietary and different" software. The end result is that there is no more proprietary Unix, and SCO goes down the drain.
Um, you cheated.
MSWindows2000 was released on February 17, 2000.
RedHat 9.0 was released on March 31, 2003 (All release dates for RedHat are from this link.)
So Redhat had over 3 years to fix holes in the distribution, while crackers had 3 more years to find holes in MSWindows2000. If you want to play fair,
- use Windows2003 released April 24, 2003 and RedHat9.0 (24 days between the releases), or
- use RedHat 6.2 (released March 8, 2000) (19 days difference from MSWindows2000) or maybe RedHat 7.0 (released August 28, 2000).
If you want to compare RedHat9.0 and MSWindows2000 today, you should fully patch both of them.
I believe RedHat will still win any of the fair tests, but if you use unpatched RH6.2, it will get cracked. It will just take longer because there are tons of script kiddies just sending easy URL cracks at every web server hoping it is running MSWindows.
My mistake, bad example: I forgot the Apache License is BSD-style.
Your mistake: bad example: MSWindows is not GPL. Their licenses are extremely draconian. Anybody releasing their source code would have very serious legal difficulties. But as a consultant, companies often share their proprietary licenses with me so I can work on their systems. As long as someone pays for the licenses, the proprietary software companies do not care whether the license is used by an actual employee or a partner. NDAs do not affect this practice, just financial wisdom.
You are correct: Patent law does not apply here. (For the record, our software is completely home-grown, and does not include code from any other project, GPL'd or not.)
The companies in question engaging in the NDA covered distribution changed the NDA's and published source after chats with FSF's legal counsel, even if they were just temporarily engaging in the practice for understandable reasons.
Do you have links? I do not remember this method being used. I remember several cases where GPL'd software was distributed, and the FSF was able to get the source distributed too. And you state that the NDAs needed to be changed, so maybe they were legal, but did not keep the spirit of the GPL? Most companies want to keep on the good side of the open source community, and will bend to keep them happy. The good publicity is worth more than any advantage from keeping the code secret for a few months.
This thread was speculating about a possible reason (losing the Unix license) that IBM would consciously break the spirit of the GPL. If IBM really needed an Unix-like OS in a hurry, they would probably start with a mature BSD rather than the weaker and GPL-encumbered Linux. (That is what Apple did.)
As I said, IANAL. I am interested in the links that this has been tried. But I seriously doubt IBM would bother trying it.
There are several good reasons for MS to delay Longhorn. (Please remember that I believe MS will enter its death spiral within 2 years.)
The first reason is financial tactics. MS has managed to get companies to pay for 3-year licenses. They have already paid, so why give them anything? MS will release the next OS at the moment just after most of these licenses need to renewed. "Renew now and the next OS is included." generates much more income than "You just received everything we expect to release for the next couple of years. Would you like to renew?"
The second reason is publicity. Talking about Longhorn puts MS in the news, which always helps the stock price. As long as it is vaporware, they can do a press release anytime there is not a new virus threat in the news. Releasing the software slows the PR.
The third reason is their business operations. They are reducing the number of programmers in the U.S. and hiring programmers in India to do the work. The Indian programmers probably keep saying things like: "This code you sent us does not make sense. There is no way it could ever run, and if it did, it would cause memory leaks and crash anyway." Moving code production overseas always incurs some delays, but it will take a long time for anybody outside Redmond's groupthink to understand this code.
The final reason is that Bill knows his company is declining. Why spend the money to develop software when your company will not exist to release it?
I always thought Pascal was another MS-related casualty.
In the 80s, everybody was talking about Pascal being the next great language. There were already descendants like Modula2 and ADA that were gaining market share. (This upset me because I learned Pascal, then C, and C looked more logical, so I hoped it would win.)
Then MS introduced QuickBasic, which turned into VisualBasic. (I have a book published by IBM stating "VisualBasic is a trademark of IBM.") Around 1992, you could probably translate programs from Pascal to VB or v.v. by substituting the keywords. The structure and the syntax was almost identical.
VB became the language of choice for beginners. It filled the same ecological niche as Pascal, so Pascal withered.
---
The difference between Pascal-based languages and C-based languages:
Pascal-based languages allow "procedures". Procedures can define multiple inputs and multiple return values. VB and the Windows API follow this syntax. It is easier for beginners to understand, but maintenance is difficult because the programmer needs to remember which variables can be changed safely and which will be returned.
C-based languages insist everything is a function. Functions have 0 or 1 return value.
When the C algorithm for returning multiple values by passing a pointer to a struct became difficult for maintenance, they invented classes, which allowed the multiple values to be retrieved. That was C++, but most early tutorials explained it as an addition to C, rather than a move to OOP. This is still evident in the ease of mixing C and C++ code in the same program.
Then Java added easier memory management and notification of errors (and reintroduced the virtual machine for sandboxing.) It follows the C++ model, but prefers OOP design.
SCO has killed Unix-branded Unix. Can you imagine any company willing to buy a license after SCO started suing some of the licensees (IBM, SGI)? The effort to start a new OS from scratch is enormous. (See BeOS.) Any new OS is going to be based on BSD or Linux. (See Apple.)
There will never be a new SystemV-based Unix, which means SCO can never sell another license. And most of the current licensees (not including MS and SUN) paid AT&T in full for perpetual licenses, so there is little revenue from the current licensees.
Lawsuits as a business model is all that SCO has left.
limiting the receivers' rights is not acceptable under the GPL. Section 6: "You may not impose any further restrictions on the recipients' exercise of the rights granted herein.". This applies to any kind of limitations, ranging from patents to NDA's.
There is no spoon. (Sorry.)
There is no receiver. There is no recipient. The nice part (from a legal perspective) is that an NDA can make a company a development partner which is part of the company for purposes such as patents.
So my company rewrites half of Apache. We do not want to distribute the code, but you are a friend, so we will lend you a server with everything configured. You ask about the source code, and we present the NDA stating that you are a development partner, will be testing the software in a production environment, and have no rights to any of the modifications.
Under U.S. patent law, that is legal, and there is no distribution of software because those companies are partners. So there is no "recipient" to have rights restricted. So there is still a hole in the GPL.
---
I like the GPL. I think it is as good as it gets for sharing software.
And this hole cannot be closed. Try adding something stating the moment you change something you are required to tell the world. Who would touch software if you had to issue a press release every time you changed a line of code? You must be able to work on it without being required to distribute.
(IANAL. Is this a good place to put that?)
--- Why I know this
We are developing software. We think we can get patents. We cannot discuss the features that might be patented with anybody with whom we do not have an NDA without losing our ability to patent those features. Well, we would have one year to file in the U.S., but we would immediately lose the right to file in Europe. The NDAs protect our rights to file, and we insist all potential customers sign them before we give a sales presentation. This limits to whom we can sell, so we want to get the patents filed so we can skip this step. It is possible for a company with the clout of IBM to force every "customer" to sign an NDA to become a "partner" to avoid the "distribute" clause.
Again, I am not saying this would ever happen, just that it is possible.
Have you even read the GPL? Any modifications to the Linux kernel are necessarily GPL'd as well.
Have you even read the GPL? Did you actually read my post to which you are responding? The GPL has a very narrow scope that is easily avoided.
The paragraph in my post just before the one you criticized explains the loophole in the GPL. Yes, modifications to GPL'd software are GPL'd. But the GPL only applies when you distribute software. If you make changes to Apache, and only use it at your company, the software has not been "distributed" and you are not required to provide the source to anybody. If you give or sell it to another company, you must make the source available to that company, but you could contract that they not distribute the software or the source to anybody else. An NDA (Non-Disclosure Agreement) allows you to share technology without allowing that technology to be distributed to others. A well-written NDA could be used to sell GPL'd code without providing source code.
Most software comes with contracts/licenses that state that it may be used on a certain number of machines for a given time period and may not be shared. All the DRM news is about companies trying to enforce those contracts through technology, even though the technology does not allow for fair-use.
Even the SCO circus is about the same battle. SCO thinks they own the rights to Unix, and they believe that the licenses to sell Unix implies that the licensees must protect Unix. Given that, then any code written to improve Unix cannot be shared with OSes that compete with Unix. Yes, they have warped minds, but it does make sense if you take the correct drugs.
I was writing about a hypothetical future where IBM dumps all of their technology that makes AIX special into a Linux-based OS. It is likely that much of it can be handled by proprietary programs outside the kernel, but anything related to making the kernel robust on 64+ processors would require kernel modifications. Nothing I said implies that IBM has not obeyed the spirit of the GPL so far, but in this hypothetical future situation, they would be giving away much of their private technology.
It is possible for IBM to take Linux and turn it into the next version of AIX. My point was that if they did this, they would want to protect their technology, and there is a way for them to do it. I was not implying that IBM would violate the spirit of the GPL. I was not implying that these events would ever happen, since that requires IBM's Unix license to be revoked and I doubt that even our courts would allow that. IBM sells hardware; the OS is just an extra, so I doubt they would even care, except for the existence of HP and SGI and the other Unix sellers who would not mind reading the source to an OS that included all of IBM's tricks.
Take this as completely theoretical. I believe SCO is evil and IBM is going to destroy them. But...
What if SCO is able to revoke IBM's license to sell Unix? The contract is broken. IBM pays $3B. And IBM needs a quick replacement to convert all those AIX installations.
The contract is broken, so IBM no longer has any obligation to protect Unix. And the customers demand all IBM proprietary technology must be in the replacement that the courts have ordered IBM to provide.
IBM cannot use Unix. They need an OS that can replace AIX, meaning it can run their customers' Unix applications. Can this be done with OS/390? OS/400? OS/2? Or will IBM immediately pump all that technology into Linux?
IBM Linux could become suitable for big iron in a very short time.
Not that all that technology must be added to RedHat or SuSE. The big loophole in the GPL is that it only matters if you distribute the binaries, and that "distribute" means customers, and U.S. law uses NDAs to change a customer into a development partner that is legally associated with the developer so the GPL is not relevant. But it would only take one customer to demand a replacement without signing an NDA to force IBM to GPL all this code. Will any company chance their information systems becoming obsolete because they did not sign a contract that does not affect themselves while IBM is offering a big discount?
Let us assume that IBM plays nice and GPLs all changes to the Linux kernel. They could still keep the code that uses those changes proprietary, but the kernel is improved. (And IBM offers Torvalds $1B to oversee the project, and he would take it because it would mean fantastic progress for GPL'd Linux.)
I have no idea how this situation helps SCO. I think it would be the final straw and "Unix is dead" would appear everywhere. I doubt SCO has put any thought into the consequences of their actions, other than they might get $3,000,000,000. I know that much money would make me happy, so maybe that is the only point, and they expect Unix to die anyway.
Thank you. You saved me some research time.
Yes, the Notes client has always been graphical, even in the 80s. The screen shots you linked do not express the power of Notes in a corporate environment. Today I prefer it for rapid development for web applications, although the client is still useful. Notes has the ability to consistently deploy secure and distributed applications with less than one week from idea to use.
I am designing a new platform, but I am starting by not trying to build a direct replacement for Notes. The new platform will have a superset of the features of Notes. We can write converters that can move Notes applications to the new system, rather than build something that uses the Notes file formats directly.
I sometimes hope to find an open source project that can be used as the base for the new system. From the text list of functions, Notes does everything Citadel does, so Citadel would have filled many of the functions needed for the replacement. I accept that the architecture is too different.
To the ACs who noticed that they should have been &&s, not ||s, you are correct.
About the lousy comments, you are also correct. Some of my code does look like this, because I would start by writing an English version, comment it out, and then insert code. The redundant comments get removed as time passes and the code is revised. My comments tend to focus on the business reasons, with a few to explain more complicated code or to warn about possible failures. I agree that "Code Complete" is a good book for programmers who want to move to category 4.
I wrote the example quickly to point out that using chained "||"s in an "if" were not necessarily bad, but if it made the code difficult to understand, then it was bad. So there was no fifth category of programmer, since someone who coded this way would still fit into the first 4 categories. (A case could be made that many who program this way would be category 1.)
About the lousy code, sorry. My first example had functions with names like "action1()", and could have done anything. There were 4 examples of code, with the last one attempting to get cute by finding a possible business use for the code. I redid the other examples to match the last version, then deleted the middle ones to save space. I had not noticed that I needed to switch from ||s to &&s to make sense. I almost never code that way, since it does make the code less understandable. Try working on code you wrote 10 years ago. You will wish it was very easy to understand. Then imagine someone else trying to work on it. I do not want the negative karma from having that other programmer curse at me.
Your fifth type of programmer fits in either the third or fourth category.
0 ));i++) notifyManager(x,z);
//Give raises until someone does not deserve one or an error occurs( record))>0) //get next employee //get number of days, errors return 0 //get amount of raise, verify it is positive number //Next record //tell the managers so they can verify before telling the employee.
3. Those who write good code most of the time, but who don't fully realize their limitations
If you write code that is difficult to maintain for the fun of having written it, you belong in the category of those who should be limited. Let another programmer review your code before it is committed. If any line takes more than a minute to understand, he either rewrites it or passes it back.
4. Those who really understand the language, the machine architecture, software engineering, and the application area, and who can write textbook code on a regular basis.
Part of software engineering is knowing that code will require maintenance someday. The code is easy to understand. The code itself code be complicated, but comments make certain the next programmer knows exactly what is happening.
severely abuse short-circuit evaluation
It is the difference between:
for(i=1;((x=getEmployeeNumber(i))>0)||((y= getDaysSinceRaise(x))>0)||((z=calculateRaise(y))>
And
for(record=1;
((empnum=getEmployeeNumber
||((days=getDaysSinceRaise(empnum))>0)
||((amount=calculateRaise(empnum, days))>0));
record++)
notifyManager(empnum, amount);
This is the same code, but the latter is more understandable.
It is not the code that matters, it is how you use it.
I had never heard of Citadel. The function list sounds like Lotus Notes without rapid application development, or any development environment for that matter, although the homepage says Citadel can be used as a datastore for applications. And Citadel has native ability for accessing file systems. Thanks for the link. I may install and review it.
Open Source replacements for Lotus Notes try to recreate all the functionality while keeping all the flaws, similar to Linux windows managers trying to become MSWindows, although at least the WMs include improvements like multiple desktops. Citadel is open source and includes most of the functionality of Lotus Notes. If someone created a development environment that followed the patterns used in LN, it might serve as a complete replacement. Of course, I have not installed Citadel yet, so it is possible that the architecture makes it completely inappropriate for this.
Do you have any experience with the Citadel developers? They ask for "small patches". Any chance they would add a major enhancement, or would it be better as a separate project?
---
About the Flamebait:
I probably should have put the first line in parantheses. I thought the parent post was quoting somebody and forgot to turn off the italics. Since it is standard to put quotes from other posts in italics, I thought a gentle reminder would be OK. I was not posting just to complain; I really expected a +1 Funny if it was modded at all.
I changed my sig because my 120th post was the first to be modded down, and I felt it was incredibly funny that the post was modded "flamebait" while talking about sun power. OTOH, I have been getting moderator points 2 or 3 times per week, and I have not received any mod points since that post. Or it could just be that the Slashdot system has been changed, since the "Quote of the Hour" has not changed in 2 days.
If I had a TV, I would want to tell my television to get the current streamed content from:
i n the near future.
tv://hbo1.hbo.com
The television has my logon, and uses it to access the content. HBO gets my IPAddress and logon, and can be certain that I am only receiving the content at one location at a time. They could track exactly what I watch, but that is a different issue. I could pay HBO by the hour, or a monthly fee.
The cable, telephone, and ISP companies should not be involved in the transaction. Why should these infrastructure companies have anything to do with my content?
Because they own the wires? I do not want a monopoly to have any control over my content. Yes, they made the investment. Yes, they should make their money back, charge maintenance costs, and profit for as long as wires are still useful. But let me see what I want to see, not what they decide to show.
I am using HBO as my example because they have shows that I would watch, if I could get just those shows. HBO could charge up to $1/hour or $20 per month, and let the first episode of each new series be free. I saw "The Sopranos" and "Six Feet Under" a few years ago. I would pay $8 per month to receive those 2 shows for a few months of each year (if I could watch them when I want. I still would not plan to be home at a certain time just for entertainment.) But I will not pay for tons of content that I do not want tempting me.
If this happened, there would probably be
tv://sopranos.hbo.com/season3/episode5.show
I doubt anybody in the industry wants this to happen. New shows would be much more difficult to push. And shows with loyal followings like ST:TOS would be harder to cancel when exact numbers of watchers is known, and revenue can be attributed directly to each watcher. Shows with limited audiences could survive by basing their production budget on the revenues. Amature production companies would have a chance.
The cable companies would still exist as both the infrastructure (probably killing the telcos) and financial centers for collecting subscriptions, but the public would have more control.
Why shouldn't infrastructure companies make money? They made the investment; they should get the returns.
The cable companies could still negotiate with the content providers for discounts and bundle everything and act as a clearing house for subcriptions. Unless they decide they do not want to carry the WB or that amature station. Then the "subscribers" could buy the programs directly from the producers. Most people would prefer to have a single bill for most of their television channels, and the discount for bundling the usual channels would be a considerable savings. But this way the subscribers have choices.
Thank you. I never took the time to see if anybody else was using the format. I used it. People understood it. It was better than what I was taught. I trained myself to always use it.
Around 1996 I was marking computer file names (specifically my resume and backups of programs in development) with the date. I switched to year first so they would sort correctly. I was working on international projects. Once I realized other people had no problems interpreting them, I started using them everywhere. In early 1999, I started using a 4-digit year to avoid the Y2K issue, and I have not written a digit-only date in any other format since then. (My resume's cover letter does contain the "Month D, YYYY" at the top.)
Using this format does generate converstaion. People usually start with "That's different, but logical. Does everybody understand it?" Now I can say it is a standard and point to ISO 8601. Thank you.
---
To the grandparent post, I am in the US, but I started using this format because I was working internationally. People in the US assume MM-DD-YYYY, but several of my programs are used globally, and the use of this format saves much confusion. I convinced management that this format would be good when we had a major issue about a deadline of 4-6-2000 involving some Europeans.
Phone, cable, power, whatever.
Separation of hardware and software and content.
(like the separation of church and state.)
Infrastructure companies made the investment to install the wires to an area. Let them make their money from renting those wires to other companies that provide services.
The "Cable" companies are winning because they laid the wires for the sole purpose of providing content, and were regulated as content providers. Now they are branching out, but want to keep the laws as they were when they were a content provider that had to create its own infrastructure. They should be considered an infrastructure company that provides content, or better, be forced to separate the functions into distinct financial entities.
The power companies went through the same issues in the last decade. In my area (PA), you can buy your power from several companies, and PECO is responsible for its delivery. The delivery charges are more than the power generation charges, but at least you have some choice.
Let the infrastructure company charge standard rates to all who want to use them, including:
- the sister company that sells service, and
- the sister company that sells content.
- any company that wants to use the wires and can afford to pay the standard rates.
I have 2 choices for information wires to my house: Verizon or Comcast. Neither can keep a connection alive for a week. Verizon has been forced to allow others to sell DSL service over their lines. Comcast has not. I prefer cable, which means I have only one choice for my ISP. It would be better if each functon was separate:
- one company for the physical connection.
- choice of companies for the software connection.
- many choices for service (mail, website hosting) and content providers (news, search sites, CableTV)
HBO could run its own service. You could subscribe to HBO, SF, and UPN without getting QVC and CBS. (I do not watch any television, so I may not have the correct channels for cool vs. uncool.)
Let the government regulate and monitor any company that owns the wires to make certain they are priced appropriately.