We use OpenBSD exclusively for our web servers. We moved our database servers from OpenBSD to Linux. I look foward to migrating our systems (some 2.8, some 2.9, one that I broke trying to do a fancy 2.8->2.9 upgrade...) when our CDs arrived. We figured that we use OpenBSD a lot, owning a bunch of CDs was worth it. Alas, it is is still cheaper than the copies of RedHat that we pick up.
OpenBSD has a real problem that I was never able to resolve, this makes it worthless for a database server. The machine is quite "efficient" with memory, which let it run with very little memory. However, with a lot of memory (our db servers normally have 1.5GB -> 2GB, I LIKE giving PostgreSQL lots of buffers and sortmem) there is little documentation on tweaking the system. I even contacted the developers in charge of the SysV memory support, etc., and they thought I hit the crack rock a little to hard.
For web servers, however, I'm quite comfortable with our OpenBSD servers sitting open on the Internet. I'm terrified of a RedHat box not being hidden. As a result, I keep the database nice and hidden.
Linux blows OpenBSD's performance away. This is a matter of Linux focusing on performance. However, for web servers (that for us just run PHP, mod_rewrite, and some other toys) I don't care... When I need more web serving power, I buy another web server for $2K. Having SSL built in to Apache is nice, and the ports is too fucking slick.
BTW: OpenBSD seems to run quite nicely on my Penguin Computing 1U servers...:)
Alex
I expect to keep our production servers on 2.9 for 2-3 months, but move development to 3.0.
There is (was) money in hosting. People need access to the Internet to send data. You can warehouse your servers or you can rent thick-pipes (T1+) that gives you bi-directional bandwidth. Therefore, hosting companies buy large amounts of bandwidth (bidirectional) or are big enough to carry it themselves with peering.
Now home users want downstream bandwidth.
Solution? Buy the bulk bandwidth, and sell the upstream via hosting and the downstream via broadband.
It's not a rude situation.
If you want bidirectional bandwidth, you can get it. Get a T1 or SDSL at home.
It costs more?
Of course it does! Upstream bandwidth is expensive, downstream is cheap.
Therefore, ADSL is priced based upon the little bit of upstream used and you get a high speed downstream connection.
It's economics. If you want upstream bandwidth, buy it. You aren't entitled to it.
If there aren't people interested, your best bet is to try to come up with a way to generate interest. Setup a set of web pages describing it, and submit it to the search engines. Place the code under the GPL or BSDL, and hope that people take an interest. Ask them to e-mail you, as you are looking for a maintainer.
However, as the code is Free, anyone can take it and use it. It appears that you are looking for free labor to do your bidding. Sorry, the world doesn't work that way. You can close up your code and it dies, or you can put it out there and hope that someone will do something with it.
With the BSD license, someone may take your code and use it, even if in a non-free capacity. With the GPL, they may use it but only in a free capacity.
You aren't interested, so move on. If you want others to benefit from your work, make it easy to find (properly built web pages to search engines can find it) and place it out for the world.
Maybe someone will use it, maybe not. Maybe they'll e-mail you questions, maybe not.
If you're done however, accept it and move on.
If there was a large team of coders working on the project, than this question makes sense. If you were providing genuine leadership, it makes sense to find a replacement.
However, they appear to be software projects that you are done with. Put the code up there. People can use it, or not. People should download it, decide where to go, and setup a fork. If you are lucky, 2-3 projects will form using your code. If not, none will.
Regardless, there is no maintainer/leadership issue, as these are solo projects.
And getting Amazon.com to understand this would be quite a trick!
Yeah, the states need to get their act together if they want to start collecting sales tax on interstate commerce that don't involve a local retailler.
Most states with sales taxes also have use taxes (alluded to elsewhere). You aren't exempt from the tax by paying out of state.
HOWEVER, the SCOTUS (Supreme Court of the United States) ruled that states can only force a company to collect the taxes for them if they maintain a nexus of operations in the state.
If the company doesn't have a presence in the state, they don't need to act on behalf of the state and collect taxes.
This is why companies can't (legally) set up subsidiaries in two states to avoid taxes. Otherwise, locals could order from another state.
The enforcement problem is that they CAN'T enforce it. They can't cross state lines with their taxes.
The Congress and Governors were trying to come up with a solution for a simplified tax system. The idea would be to at least standardize to the point where given a zipcode, a simple lookup would determine the tax base.
Keep in mind, not only do states collect sales tax, some counties and cities add them as well. This creates a mess. It is one thing to have to do a lookup on 50 states, it is another to have to deal with localities.
Companies with solutions have tried to find beta testers, but who will volunteer to collect sales tax just to beta test software that will make it mandatory.
Interestingly, New Hampshire doesn't charge sales tax on liquor (or anything, if I recall), so Mass got annoyed that residents would cross state lines to purchase things, including liquor at the New Hampshire State liquor stores (can only buy booze in New Hampshire at state run liquor stores, right along the highway... isn't that entrampment?). Mass sent staties into New Hampshire, calling back license plates, and arresting people crossing the line (or something similar)... so New Hampshire deployed their troopers to arrest the Mass employees on silly charges, and the situation went away.
States' Rights matter outside the northeast, because the states are huge and do their own thing. States' Rights don't matter in the northeast because the states like to squabble with each other and would like to have more central control because people cross the lines regularly.
The GPL was created to advance a political ideology. The GNU project was an OS that sat within that framework.
Most of the significant projects were CREATED by University projects. Somehow it is easier to create great "free" software when your developers are reasonably paid University professors, grad students, and a bunch of undergrads on "research" projects.
If your core developers are all well educated, formally trained, engineers, it is easier to avoid the differing levels of experience/documentation that amateur development gets.
Corporate development needs to produce revenue. Sure your admin/IT staff may find that adapted Free software is cheaper than either developing internal apps or buying "off-the-shelf" software, and maybe even submitting the patches.
A large majority of the "open source" software is produced by individuals that are in high school or college producing solo projects. It is wonderful that they release their software, it doesn't lend itself towards a large coordinated project. A tenured professor overseeing 15-20 students working on a project provides continuity that a student coding (who will find himself in a job in a few years, maybe even with friends or family taking up his free time) doesn't have.
University research projects are HARD to reconcile with the GPL. The BSD license (and similar licenses, like the MIT license) are great for universities. Universities producing money with corporate or government grants should produce truly "free" software, that are free for all to use as they see fit.
All Slashdot bull aside about keeping it free, BSD licensed code is FREE. Anyone can use it, nobody can lock it up. It's there, forever. If it is a university project, it will likely be always hosted, as opposed to small shops that can change direction/go under at any time.
GNU set out to create a Unix-like OS. The kernel got little attention, as the Linux kernel took off. Producing BIG applications isn't within the GNU project (and its programming offices at the MIT LCS (or AI lab, I forget where they are stuck).
End user applications require a different set of requirements.
The other problem with some of the tasks is that they are hard to divide the costs. Apache was spread among people that needed a web server, now the rest of us users leach off their work.
MS Office's costs are spread among all corporate America, with a juicy profit flowing to Microsoft. Creating a competing office suite is hard to manage, because spreading the costs is HARD. Licensing fees "spreads" the costs, but there is a free rider problem. Each Fortune 1000 would probably benefit by sending 10% of their Office licensing amount to an Open Office consortium as a long term 3-5 year investment in reducing their costs significantly, but each corporate board would prefer that every OTHER company send the money and they keep it.
Its the prisoner's dilemma, and its hard to fight.
Figure out a way to divide the costs, and you'll get quality GPL software.
Niche markets are easier, get all the vendors to line up for one solution. Mainstream markets are where "shrink wrap" software excels... there is a reason for that.
Right... because I can rebuild my Win2K desktops with my Compaq restart disk in 30 minutes, then do a network install of Office 2K. The developer can then install his favorite tools from the share drive in 30 minutes. Apply SP2 and hotfixes whenever they get around for it (before they leave for lunch).
If I was supporting a large Office, I'd have the automated installs. I got Windows to install off it (really hard, hit F12 on bootup, answer some questions, leave it alone) but got bored dealing with the documentation. Then for fun, update your drivers, I have them on the software share sorted by OS and model... I have 3 system models, all Compaqs. I ripped out the old Dell systems and put them to use as OpenBSD boxes long ago. Support is now a case.
Now, how much time did you spend installing new Themes under KDE/GNOME?
How often do you download the latest KDE/GNOME desktop to see how it works? How often is this spent compiling? How often do you decide to mess with the drivers and garbage to get sound working on Linux to support a $20 sound card (once had someone tweak for 2 days, instead of finding a supported card and whipping out the credit card).
When do you decide that you need to go "lock down" your workstation?
What happens when I need to show you the Visio diagram for a project? Reboot into Windows? Or did you spend 2 hours playing with VM Ware to get it working right?
Hell, I love my OS X workstation, but quite frankly, but Win2K laptop kicks ass as a client machine. I have network shares made available offline. This means that I can work on a client proposal on the train to/from work, do accounting work in Quickbooks, etc.
My Blackberry syncs in nicely, keeping my appointments and telling the server when it is hooked in.
I LOVE *nix. We build applications on OpenBSD/Linux. Both are great OSes. OpenBSD is the FASTEST server installation I've ever seen... My Penguin computing boxes are up running 2.9 in 30 minutes, then I just scp over the files that we've tweaked.
But you can't top the MS desktop for productivity applications. Sure the Mac GUI makes it counter intuitive, but I can deal. My software runs. All my hardware is supported.
I never lose someone for a few hours while they tweak their kernel. The only kernels that got compiled was the OpenBSD kernel. As I standardized that hardware, I scp it on to the new box and move on.
MS makes some good products. I stay off their server stuff because once I choose one, I have to choose them all. That's annoying. However, their desktops are more productive and faster. The WinXP test looks good too. The taskbar changes that they stole from OS X makes our virtual desktop program unnecessary.
If we can't access our files, we can (and have lost) $10,000-$30,000 projects.
I'll keep cutting MS checks. However, for a machine that NEEDs to be up, a Unix solution is the only way to go. Linux and OpenBSD seem to pair up nicely for us as a database and web server combo. They have great uptime, and no registry connections. However, for the desktops, I keep all the data on backed up servers, and if a machine dies, its rebuilt exactly how we want in a matter of hours. Meanwhile, with roaming profiles, we grab a spare machine while it is being rebuilt. It's a small price to pay to be in business.
Its really easy, you try to keep the execs out of the decision process. Get your requirements and develop a strategy.
Go buy the high-end Red Hat distribution. We did this for a client that knows we were using OpenBSD off a $30 CD. We also buy the nice Red Hat distribution.
Call up Red Hat and set up a support arrangement. They are a reasonable sized company. Alternatively, call IBM.
The advantage with Microsoft is that if you are a big company (say, the Fortune 1000 in the article) you get the special phone line for support, etc. Microsoft supports the Fortune 1000 a little differently than your pirated copies of Win95.
Don't sell them on free. Tell them that you worked with Red Hat's OS, and you found that it is better suited to this project. Inform them that you can reduce downtime (their real concern) instead of a couple grand on licenses.
More importantly, emphasize that it will save you time.
We don't use Linux or *BSD on desktops, it is too expensive.
Win2K or even WinXP involves a short installation procedure (before lunch) then come back and finish. Setting up a Linux desktop (for a technology guy, not end users) takes a few days of playing. Win2K tweaking with fun apps takes $1000 in software (including a $200 shareware budget) and you're good to go.
Look at your salaries. See what it costs your company to have you putzing around for days.
BTW: with the MS licenses and a point-and-click installer, how much time does it take to get another server up and running. Including your downloading Redhat over the corporate T1 (or whatever), how much of your time is spent putzing around on IRC, etc.
Sure, IRC is nice for REALLY hard problems. However, having a server down for 1-2 days while you troll USENET or IRC for help isn't acceptable.
Next time a MS solution is being proposed, try to get 24 hours to stall. Take the same list of software, and the budget, and CALL Red Hat Sales. Tell them what is going on, and ask them to put in a bid.
Alternative, call a Linux consultant, and work with them to put in a bid for the implementation AND for the Red Hat support contract. If the Red Hat fee is less, show that to management.
You all would get a LOT more credibility with management if:
A) You dress like professionals (I did NT Consulting for 4 years... we all wore a nice shirt and khakis... the Linux guys would often wear jeans, it makes a difference; my BSD shop does it too, it matters)
B) Emphasize solutions, not technology (they are looking for a solution, show that you understand this. Emphasize the savings in downtime, not licensing fees.
C) Focus on REAL cost savings. Don't CONSIDER unsupported downloaded applications. Discuss support agreements, Red Hat Network, etc.
Geeze, this isn't rocket science guys, understand what the executive is trying to accomplish.
Alex
Monopolizing the set-top market
on
Gamecube Guts
·
· Score: 2
Yeah, my bad, that is what I meant by the console market. Gaming is the first foot in, I would expect a follow-on product RSN (real soon now) that includes a large hard drive and TV tuner. It would be a UltimateTV/XBox combo with the DVD player. They might even bundle an MP3 player.
Once they have the system, all these other uses require no additional processor, etc. They can embed the equivalent of all these add-ons.
They need a strong gaming market to do this however. Why?
The processor/graphics are best subsidized by the gaming market.
Think about it, if there is a $300 price for the gaming, they can add the functionality for little more. They have the $300 Xbox, and several $400-$750 Xbox+ systems.
MS is always about bundling. The audio/video market is split into a low-end integrated solution and a high-end component solution.
MS will target the low-end (their specialty) with the integrated solution, and Xbox will be the first step. I would expect in 3 years Xbox2 which plays Xbox games as well as some new Xbox2 games. They'll speed up the process of console replacement, but game makers will just set whatever requirement they like. No problem that Xbox5 is out, some games will require Xbox1, Xbox2, Xbox4, etc.
The real trick for them, IMO, is getting the gaming market to make people pay for the processor. Once the processor/graphics is paid for, the extra features are just a matter of software and harddrive space, neither of which bumps up the product much.
Xbox is a console, designed to make money off games.
The long term goal is the monopolization of information into the home. Then they get a cut of EVERYTHING: games, music, video, time-shifting, etc.
As a whole-widget company, it will be increasingly difficult for single solution players to compete. Sure the high-end will never adopt the MS all-in-one solution, but Panasonic, AIWA, and everyone else that plays in the space is in trouble.
Expect Xbox based solutions to come in all forms... including those with a built-in amp to power 4, 5, or 6 speaker configurations. Some will include speakers, some won't. They'll create a family of solutions that share the same core and come in different bundles.
That's scary, if only because of their previous licensing strategy.
Alex
Yeah, there Xbox losses are irrelevant
on
Gamecube Guts
·
· Score: 2
MS wants to sell the whole widget. Dell has profits. People buy Dells (generally) to run MS OSes. As a monopoly, MS is able to extract MOST of the profits. If they were a pure monopoly, they would get it all. MS has SOME competition, so they have to leave some profits to the OEMs.
XBox is a multi-billion dollar play to get their systems into homes. They are trying to monopolize the gaming market.
The problem is, without analyzing the console penetration, game developers won't know if Xbox sales are propped up by Slashdotters saying "damn the man" and buying them to play with.
If Xbox looks like it has great penetration (and 1 million "nerds" buying them to play with WILL be significant in the first 3 months), the games will come for the Xbox, and all us Gamecube owners will be left high-and-dry for third party apps.
Oh well, I've been reasonably happy with every Nintendo console because of the first party apps. My favorite was the NES followed by N64 followed by SNES, but I enjoyed them all.
The N64 was a failure, but I loved Mario 64, the first Zelda game, Goldeneye, Smash Brothers, and Hang Time. Those were enough games to keep me interested in the console.
I am getting a Gamecube for Blitz 2002, the new Smash Brothers, the new Hang Time (forget what this one is getting named), plus the new Mario and Zelda games. A new Metroid and Star Wars games are icing on the cake.
I haven't seen anything for Xbox that makes me want one. Halo looks nice, but I'll wait for the full featured PC version. I mean, anything good for the Xbox should be ported to the PC, and the PC will have games that the Xbox won't be good for (RPG/Strategy games that I really love). I have a HTPC for gaming in my system, so the TV isn't even an advantage for hooking up a console.
PS2 is starting to look good with some exclusive games. I may pick it up as a second system. The Gamecube just looks more impressive to me and has the games that I want.
Hackability? Give me a break. I want a gaming machine so my friends (the human kind, not on IRC) can come over and we can play a few games after work. You're right though, MS isn't going to be hurt by losses, not their style. When they have a high stock price, they use it to persue global domination. When they don't have a high enough price, they use cash to do so.
Their shareholders? Management controls enough of the shares to avoid a take-over.
Taking MS on head-on is likely a failure. Your best bet to hurt them is to create an alternative in either core or secondary markets. Buying a Gamecube helps fight MS taking over the console market. Buying Linux servers helps stop NT's spread. Keep MS contained, build alternatives.
The only reason to target MS is if you can find a way to stop them from coercing you. I'm not concerned that they make a lot of money. I'm concerned if they can dictate the Internet on the server side because of a client-side monopoly.
Microsoft has a monopoly on desktop OSes on the x86 platform. This results in games having to adapt to Microsoft APIs to compete.
Therefore, all PC games had to support the APIs. Microsoft is clearly leveraging its monopoly on the desktop to establish a monopoly in the console market.
Sure, the X-Box games won't pop into a PC, but you can probably share between 95% and 100% of the code between the versions.
I think that the dumping arguement is stronger, but the shared APIs help the case.
If the XBox truly costs $375 in marginal costs, and they are selling it for $300, they are engaging in dumping. It is illegal to sell products under marginal cost in an attempt to monopolize the marget.
A combination of Microsoft leveraging Windows (the DirectX API), a monopoly, into establishing a new monopoly through market dumping, should make this a no brainer.
MS got away with IE on a technicality. Because software, as we all know, costs 0 to copy, MS was able to give IE away without engaging in dumping because the marginal cost was 0.
They did spend bandwith. However, as bandwith is a fixed cost (you pay $X for Y Mb/s), the marginal cost for the copies remains zero.
I don't know if the idea of selling below marginal cost to make up backend revenues matters, but the entire Microsoft practice smells of dumping.
They are using their cash reserves and selling below cost in an attempt to put Nintendo out of business (Nintendo is a games company!) and cause Sony to bleed red ink.
This seems like a plain approach of market dumping.
The interesting thing is, most publically traded corporations are stereotyped at looking at the quarterly returns. Microsoft Corporation is unique in that it DOES look out for the big picture. The successfully leverage their monopolies and cash reserves to "cut off the air supply" of their competitors.
If Microsoft is selling under marginal costs to establish marketshare and monopolize the market (where they can then jack up the licensing fees, etc.) this should be clear cut.
Are any of the lawyers on Slashdot around? Am I on the right track?
Everytime a thread on Slashdot takes place about gaming consoles, 5-6 people get scores of 3-5 (insightful) for pointing out that game makes lose money on the hardware to sell the software.
This gets mentioned frequently, and 3DO is mentioned as an example of what happens when they don't do that.
I've always been wondering, do we really know that this is true? Has anyone any financial data to prove this?
If there is some proof, can the Slashdot editors include this information in each posting, so those of us browsing at higher thresholds don't see half the posts with this insightful fact?
I also don't really buy this theory. I mean, how much can the licensing fee be for each $50 game? Also, in an age of video game rentals, how many games does the average console owner own?
I mean, if you figure that the average game now sells for $50, the store pays AT MOST $35 for the game. The distributer probably picks it up for $25 (so the BIG stores get the bigger margins, no separate distributers). This leaves $25 to be split among the maker and the console. I can't imagine that the license is more than $5. Maybe it is $10? That would explain WHY Nintendo and (until recently) Sega made systems, $5-$10/game is a nice margin, plus they get the revenue for the author when they sell their own games.
Let's figure that the average console owner owns 10 games/console (that seems REALLY high BTW, I owned 30-50 NES games, but they were mostly the original $30 games, and their weren't rentals in the early NES days), plus rents enough games to result in the local store stocking an extra 10 games. This is 20 games/console, at $10/game, yielding $200 in licensing.
Now, how much of that licensing is Nintendo or Sony willing to spend subsidizing the hardware?
I had always heard that the stores make little margin on the systems (not a loss, but a trivial profit) and make their money on the games/peripherals. This makes more sense, as they trade a little bit of store space to get the margins on the games. The games are good for toy stores, as the space/product is minimal compared to real toys and the prices are high.
However, the console maker subsidzing the hardware (more than a trivial fee) seems absurd. I mean, MAYBE the launch versions get subsidized, but given the demand (preorders, unavailability for 2-3 months), why would they subsidize sales when they could clearly move the units at cost or above cost. Now, I could see subsidizing post-launch consoles to move sales, but manufacturing costs should go down over time, allowing the prices to drop (which they do) or the profits on consoles to increase.
Now, I COULD buy that the console makers sell the machines at cost. This would result in a subsidy of the "fixed" costs (R&D, setting up manufacturing process), but still, this wouldn't be real. As the costs go down (consoles stay on the market for 5 years, electronics go down tremendously in 5 years, but consoles rarely drop THAT significantly in price... i.e. a $300 console may drop to $200, but the manufacturing costs 2 years out should be half the initial costs), maybe they recover the initial subsidies?
I mean, the common Slashdot belief that the 3DO was $700 because of no subsidies and needing a profit, while the $200-$300 consoles are sold at a loss is ABSURD. That implies a $400-$500/console subsidy (or $200-$300 with an extra $200 in profit for 3DO makes), which would result in assinine losses.
In computers, processors are sold at a premium when new, but moved at lower prices after R&D is recovered and Intel was traditionally trying to move more units. This is simple price discrimination, not a value judgement on valid profits.
I mean, maybe you sell consoles at a loss to create a software market. Then the software market creates a demand for the console which allows profits on the later consoles, but this subsidized hardware "theory" makes no economic sense.
Realize that 3rd generation games are planned AFTER the launch (first generation hit with the product launch or soon after, the second generation normally starts in after the first round are finished before launch, and the third generation don't start planning until 6-12 months of sales are known), so creating a demand by moving consoles allows third generations products to be made.
However, I'd like someone to either provide EVIDENCE of this subsidy, or at a MINIMUM some economic analysis to show WHY companies would do so. This simple assertion ("remember, consoles are sold at a loss") is neither insightful or useful.
Alex
P.S. Apologies for singling out one of the posts, this could apply to any of dozens of Slashdot posts on the subject.
Right now, my DSL "modem" and Firewall/WAP sit in my study with one of the computer, and we use the wireless for the laptops. (I live in an 700 sq. ft. apartment, so for a home, YMMV) The plan is to move the DSL/Firewall into the entertainment center, which happens to be in the middle of the apartment because of the layout.
Originally, I was going to use 802.11b to get connectivity to the HTPC, etc., but with the growing availability of ethernet support in devices, I concluded that I should have a non 802.11b connection in the entertainment center. Their already is a phone line run back there to support the Replay, so splitting it and running the DSL there is trivial, and I will eventually have 3 devices in the system that need connectivity.
The Gamecube is going to support an ethernet addon, some sort of MP3 player may make it into my system (sure the HTPC - home theater pc - could do it, but it creates a UI nightmare... although having a Pronto makes life easier), and the HTPC obviously needs connectivity. I concluded that the USB-802.11b connection will be adequate for the docking station that's for the study, as well as the second computer.
Yeah, my freaky apartment has a docking station for my work computer, my fiancee's old PC that she uses for her Windows apps, my HTPC for computer gaming on the television, my laptop out of the docking station, and the fiancee's iBook. I guess it is a problem that I love technology for technology's sake, oh well.
The problem with 802.11b is that unless you have an intelligence antennae system, reception blows. My laptop whines at places that the iBook is fine at. Apple's engineering is impressive.
802.11b is amazing though, and more stuff should support it. I would love to get some of the LCD-based computers if I could get a good 802.11b setup, so I could display rotating digital photographs throughout the apartment. Oh well, one day.
Alex
Check the logs...
on
Netscape 6.2
·
· Score: 3, Interesting
No doubt, coding for IE is cheaper. As I posted in another thread, one of our clients made that decision, and now we have to rebuild the site to be Netscape friendly? Why? 10% of his "unique visitors" are Netscape, and they can't even use the site with the latest version.
If a 10% increase in profits > cost of implementing a Netscape version... well, Netscape version is coming...
Its a business decision. The IE5 version of the page is the low hanging fruit. Netscape is more of a challenge... Now if I could figure out the random Mozilla rendering problem...
Not a business problem, I'd just like to see it work under Mozilla/Netscape 6.x.
Alex
NT Started at NT 3.1
on
MS DOS: A Eulogy
·
· Score: 5, Informative
Come on, don't you remember all the OS/2 vs. NT 3.1 articles when NT 3.1 shipped? NT 3.1 was a flop, mostly used as a testing ground for people interested in keeping up with MS's new plans.
NT 3.51 was the first successful version of NT. NT 3.51 SP 5 was amazingly stable... it would be interesting to put an NT 3.51 SP 5 machine up against a Windows 2000 SP 2 (NT 5 SP 2) machine and compare.
Win32s was the backwards port of the core of the Win32 API to Win3.1. The two goals were:
1) Get new applications written against the Win32 API so NT (the future) would have some applications
2) Break OS/2 Windows compatibility layer... they kept changing Win32s until they broke OS/2, then they released apps for Win32s.
Windows 4.0 (Chicago AKA Windows 93 AKA Windows 95) was the version that combined DOS/Windows (to stop the DR-DOS onslaught) and introduced the Win32 API as the standard API. Win95 resulted in the Win32 apps that allowed NT to show some success on the desktop. NT 3.51 had some success as a server (very useful environment for managing Win3.1 desktops without the cost of Novell).
Win95 had some new APIs, which were mostly ported to NT 4 (except DirectX > 3 APIs). When I was at Citrix (MS Blocked WinFrame 2.0, then basically bought it to become Terminal Server), we couldn't support newer versions of IE because WinFrame 1.x was based upon NT 3.51, and IE required Win95/NT4 APIs.
Cairo was supposed to be the end of Windows with NT 4. Two years late and without a lot of functionality, NT 4 had (and still has!) some good server-side support and corporate desktop standing. When NT 4 lacked a lot of the functionality, MS declared that Cairo was a set of projects, not a release, and that some of them would be in NT 5. NT 5, two years late as Windows 2000, finally made a nearly API complete NT to match their home desktop dominance.
Windows XP appears to use a nearly identical system, focusing on a new user experience based on MacOS's improvements.
Microsoft has finaly achieved its 8 year goal of eliminating DOS support, ME was the end of the DOS based Windows, and it looks like all the old DOS games are finally dead. MS kept promissing better support for DOS apps/games in the next version of NT, but never delivered, instead stalling on their demise. Oh well.
Interestingly, NT 3.51 (I don't recall NT 3.5) was extremely portable, commercially supporting 4 processor families (this continued until NT 4, but the other platforms failled to take off).
The DOS support in NT, the NT VDM, emulated a 286, albeit much faster. This is the reason that you couldn't run fancy things in the DOS emulation, if it was a protected mode DOS API (386 DOS app), the NT VDM couldn't handle it.
Hopefully a better solution than VMWare (overkill, complexity, etc.) will exist to run old DOS games in emulation. My brother bought me the commercial version of Abuse (at one time a favorite) as a present, but I got it about 2 weeks after I migrated to NT 4 fulltime. Well, my new HTPC (home theater PC, just for gaming, I got me a progressive scan DVD player already) is going to be 98SE or ME based for gaming compatibility, so I guess I'll be able to play the old classics there.
This site is almost entirely a one-hit wonder. You hit the site, sign up, and leave on your way. We're trying to focus on the users hitting our site. The standard site is CSS/HTML compliant (some IE additions, but it should gracefully degrade).
We're looking into cleanup options to make it cleaner and therefore run on Mozilla, but Netscape needs a custom solution. Maybe I should get a WebTV system and try it out, we'll see.
Sending the user away from the page means that they aren't generating revenue for my client. We're not interested in improving the web, we're interested in improving their bottom line.
If it works in my Mozilla browser, terrific, if not, oh well. If and when Mozilla/Netscape 6.x provide enough of a reason to make the site compliant, we'll work through their bugs.
It's annoying, but IE/Netscape 6 conversions should be easier. I don't mind (too much) writing two stylesheets. They don't take that long. It's making two versions of the site (a legacy one for Netscape) that is annoying me.
I test in IE because thats what the users are using. I'll develop for Netscape 6 when the platform is available.
The central codebase is the same, I just need to write different HTML renderers...
Sigh, one of our projects is to write our own XML language that was a content/display combo that wasn't HTML. Then we'll just write three renderers, IE/Netscape/Mozilla. Oh well, one day.
I spent months trying to understand the tech before making a purchase. I then spent months trying to understand how to best use it after I made a purchase.
I appreciate that I don't fully understand it, but I won't apologize. You guys that work with the technology would be really helpful if you provided information (like this post did) instead of just telling us that we are wrong.
Sit on some of the home theatre boards. The HDTV engineers on them disdain people that don't understand things as well.
It shouldn't require 4 years as an engineer at an HDTV firm to understand what you are buying... My four year MIT EECS degree should be fine...:) Hell, anyone, regardless of background, should be able to find information if they dig deep enough.
I find your arguement that interlaced is more pleasing fascinating. However, I would still rather the information be transmitted at 480p. 480p is twice the data as 480i.
The reason that I find 1080i interesting is that it is quoted as an interlaced spec. Many people think that 1080i is better than 720p for this reason. This is blatantly false. A 720p image has more data.
Ideally, you want to get as much data as possible into your system. Once it is in your system, you want to be able to customize how it displays based upon your preferences...
For example, I may want everything to display at NTSC levels through my VCR, but I still want the most data reaching my house. I (through my equipment) should be able to do what I want to display it best, but I still want the most data in.
One of my client's sites was written with just IE in mind. It makes heavy use of CSS, and Netscape's CSS bugs just cough on it.
However, the logs indicate that currently 8.5% of our users are Netscape 4.x.
The operations guy at the client broke out his calculator, saw the costs of my fixing the system for Netscape, saw the revenue/profit increase, and saw that B>A and said, do it.
I was hoping to just change the style sheet, but Netscape is totally busted, so it looks like separate scripts. Sure the IE version will be the priority, but when you can increase profits 8-10% of more (in fact, increasing revenue by 8% should increase profits 10%-12% based upon some fixed costs, etc.) it becomes really hard to justify ignoring.
Unless technology costs are a rediculously high percentage of your budget, you can't ignore 8% of the market.
Now WebTV and Mac, that are.5% and 1.5% of this website? They probably aren't worth spending resources on beyond testing on the Mac, but you have to evaluate your costs.
What about non-commercial sites? Code to HTML standards, and use minimal CSS. While we have sites that need heavy CSS to look amazing, the site could work without them. Limit yourself to fonts, sizes, etc., and you'll be fine. Don't worry about it looking right tot he pixel and you'll be fine on multiple browsers.
Okay, a standard television can do 640x480 interlaced (called 480i if you follow DTV specs). This is 60 fps interlaced, which is really 30 fps. Alternatively, you can do a 240p signal in the 480i system, which is what the original Nintendo system did. (IIRC)
With a HDTV-ready system and a Gamecube, you can do 480p, which is 640x480 progressive, which can provide a true 60 fps signal.
Because of the size and shape of my living room, I have a standard size television, which is a 4:3 ratio. Some of the HDTVs are full-screen (4:3) and some are wide-screen (16:9). You can send a 16:9 signal to a 4:3 television and it letterboxes (or you swap the aspect ratio and get tall, thin people). You can send a 4:3 signal to a widescreen and reverse letterbox (on the sides) or stretch it and get short, fat people.:)
In supporting 480p, the Gamecube offers a true 60 fps, as the entire screen updates every frame. With a standard television, running at 480i, you will really only effectively get 30fps, as it takes 2 frames to draw a full image.
Interesting, the 1080i ration for HDTV (which can also carry a 540p signal) is interlaced. Interlaced is fine for things without that much movement, games and heavy movement systems benefit from progressive images. Computers have all but abandonned interlaced signals, and with HDTV, you'll have to find a TV/STB combo that does what you want. The 720p resolution will be amazing for HDTV and the "next" generation consoles after this crop.
In high school, I used to check Tom's Hardware Guide regularly. His posting on over clocking seemed fascinating. He seemed to have a good understanding of hardware issues.
Then I went to college, still a reader.
Taking a few courses on hardware however, made reading his reviews painful.
He doesn't understand hardware, and it shows. Sure, he is able to run tests, but his reasons are completely flawed.
I don't accuse Tom of being on the take... However, I think that Tom should stick to testing, and not give his "engineering" insights that are based upon made up terminology and without an engineering basis.
This is a more reasonable review, and the first one that I viewed of his in years.
I've been evaluating Mac OS X for a few months. Our applications are primarily web based, though we do Java and are adding Qt to the technology base.
We had each programmer with a Linux box and Win2K box. Now it is hit or miss. If you need/want a dedicated server, its yours. Otherwise, you use a development server.
Mac OS X offers some interesting possibilities. One system, with a full BSD subsystem (adequate for us), as well as an Outlook Client and MS Office (coming to native OS X, "Real Soon Now"). When those are released, we're really going to consider Powerbooks instead of the Compaq Armada's with docking stations.
Giving everyone a full Unix desktop (complete with CVS client) as well as Office Productivity apps would be great.
VirtualPC would probably even be sufficient for the few Windows only apps that we need. (Test the site on IE/Windows, Quickbooks for my accounting, etc.).
I realize that if you are doing real Unix coding, you need the Sparcs. But for those of us that just need a Unix-like environment, it is a sufficient development solution.
Despite deploying and building our systems on UNIX platforms utilizing BSD/GPL systems, we have an NT 4.0 network with Exchange 5.5 running.
Why?
We need the coordination ability of Outlook. I haven't seen anything for UNIX that compares. The ability to schedule meetings, observe schedules, and otherwise coordinate things is extremely useful. I would love to replace the Exchange Server and backend infrastructure with Free Solutions, but they aren't there yet.
I use IE and Mozilla as my browsers (IE 6 is pretty flaky), but I am stuck on a Windows 2000 desktop (albeit with 3-4 SSH sessions going) because of collaboration tools, accounting tools, and office tools.
This is a step in the right direction for those looking to build Free solutions.
Jabber, Inc. (jabber.com) is certainly an obvious choice. Keep in mind that the RFP will include what you need, not the technology. Any company that works with (supporting) Jabber is a good choice.
Anyone can take the Jabber source, modify it (if needed), and install it for you. While Slashdot would probably like you to go to the authors, in looking for your corporate bottom line, you should contact ALL companies that work with Jabber.
Sure the source is "free", but who cares?
What matters in this case isn't the license (though the GPL may make the Jabber-based bids more pleasant), its the result.
Each company should submit you a proposal with a description of how they will meet your needs and a price.
All the companies that submit with Jabber have an advantage, they don't need to support the full infrastructure of development, just their portion. The Jabber, Inc. people should be able to bid competitively, given that they already have a sales and support network in place for selling Jabber.
Good luck, and thank you for considering open source solutions in your RFP process.
Slashdot Mirror
We use OpenBSD exclusively for our web servers. We moved our database servers from OpenBSD to Linux. I look foward to migrating our systems (some 2.8, some 2.9, one that I broke trying to do a fancy 2.8->2.9 upgrade...) when our CDs arrived. We figured that we use OpenBSD a lot, owning a bunch of CDs was worth it. Alas, it is is still cheaper than the copies of RedHat that we pick up.
:)
OpenBSD has a real problem that I was never able to resolve, this makes it worthless for a database server. The machine is quite "efficient" with memory, which let it run with very little memory. However, with a lot of memory (our db servers normally have 1.5GB -> 2GB, I LIKE giving PostgreSQL lots of buffers and sortmem) there is little documentation on tweaking the system. I even contacted the developers in charge of the SysV memory support, etc., and they thought I hit the crack rock a little to hard.
For web servers, however, I'm quite comfortable with our OpenBSD servers sitting open on the Internet. I'm terrified of a RedHat box not being hidden. As a result, I keep the database nice and hidden.
Linux blows OpenBSD's performance away. This is a matter of Linux focusing on performance. However, for web servers (that for us just run PHP, mod_rewrite, and some other toys) I don't care... When I need more web serving power, I buy another web server for $2K. Having SSL built in to Apache is nice, and the ports is too fucking slick.
BTW: OpenBSD seems to run quite nicely on my Penguin Computing 1U servers...
Alex
I expect to keep our production servers on 2.9 for 2-3 months, but move development to 3.0.
There is (was) money in hosting. People need access to the Internet to send data. You can warehouse your servers or you can rent thick-pipes (T1+) that gives you bi-directional bandwidth. Therefore, hosting companies buy large amounts of bandwidth (bidirectional) or are big enough to carry it themselves with peering.
Now home users want downstream bandwidth.
Solution? Buy the bulk bandwidth, and sell the upstream via hosting and the downstream via broadband.
It's not a rude situation.
If you want bidirectional bandwidth, you can get it. Get a T1 or SDSL at home.
It costs more?
Of course it does! Upstream bandwidth is expensive, downstream is cheap.
Therefore, ADSL is priced based upon the little bit of upstream used and you get a high speed downstream connection.
It's economics. If you want upstream bandwidth, buy it. You aren't entitled to it.
Alex
If there aren't people interested, your best bet is to try to come up with a way to generate interest. Setup a set of web pages describing it, and submit it to the search engines. Place the code under the GPL or BSDL, and hope that people take an interest. Ask them to e-mail you, as you are looking for a maintainer.
However, as the code is Free, anyone can take it and use it. It appears that you are looking for free labor to do your bidding. Sorry, the world doesn't work that way. You can close up your code and it dies, or you can put it out there and hope that someone will do something with it.
With the BSD license, someone may take your code and use it, even if in a non-free capacity. With the GPL, they may use it but only in a free capacity.
You aren't interested, so move on. If you want others to benefit from your work, make it easy to find (properly built web pages to search engines can find it) and place it out for the world.
Maybe someone will use it, maybe not. Maybe they'll e-mail you questions, maybe not.
If you're done however, accept it and move on.
If there was a large team of coders working on the project, than this question makes sense. If you were providing genuine leadership, it makes sense to find a replacement.
However, they appear to be software projects that you are done with. Put the code up there. People can use it, or not. People should download it, decide where to go, and setup a fork. If you are lucky, 2-3 projects will form using your code. If not, none will.
Regardless, there is no maintainer/leadership issue, as these are solo projects.
Best of luck,
Alex
And getting Amazon.com to understand this would be quite a trick!
Yeah, the states need to get their act together if they want to start collecting sales tax on interstate commerce that don't involve a local retailler.
Alex
Most states with sales taxes also have use taxes (alluded to elsewhere). You aren't exempt from the tax by paying out of state.
HOWEVER, the SCOTUS (Supreme Court of the United States) ruled that states can only force a company to collect the taxes for them if they maintain a nexus of operations in the state.
If the company doesn't have a presence in the state, they don't need to act on behalf of the state and collect taxes.
This is why companies can't (legally) set up subsidiaries in two states to avoid taxes. Otherwise, locals could order from another state.
The enforcement problem is that they CAN'T enforce it. They can't cross state lines with their taxes.
The Congress and Governors were trying to come up with a solution for a simplified tax system. The idea would be to at least standardize to the point where given a zipcode, a simple lookup would determine the tax base.
Keep in mind, not only do states collect sales tax, some counties and cities add them as well. This creates a mess. It is one thing to have to do a lookup on 50 states, it is another to have to deal with localities.
Companies with solutions have tried to find beta testers, but who will volunteer to collect sales tax just to beta test software that will make it mandatory.
Interestingly, New Hampshire doesn't charge sales tax on liquor (or anything, if I recall), so Mass got annoyed that residents would cross state lines to purchase things, including liquor at the New Hampshire State liquor stores (can only buy booze in New Hampshire at state run liquor stores, right along the highway... isn't that entrampment?). Mass sent staties into New Hampshire, calling back license plates, and arresting people crossing the line (or something similar)... so New Hampshire deployed their troopers to arrest the Mass employees on silly charges, and the situation went away.
States' Rights matter outside the northeast, because the states are huge and do their own thing. States' Rights don't matter in the northeast because the states like to squabble with each other and would like to have more central control because people cross the lines regularly.
Alex
The GPL was created to advance a political ideology. The GNU project was an OS that sat within that framework.
Most of the significant projects were CREATED by University projects. Somehow it is easier to create great "free" software when your developers are reasonably paid University professors, grad students, and a bunch of undergrads on "research" projects.
If your core developers are all well educated, formally trained, engineers, it is easier to avoid the differing levels of experience/documentation that amateur development gets.
Corporate development needs to produce revenue. Sure your admin/IT staff may find that adapted Free software is cheaper than either developing internal apps or buying "off-the-shelf" software, and maybe even submitting the patches.
A large majority of the "open source" software is produced by individuals that are in high school or college producing solo projects. It is wonderful that they release their software, it doesn't lend itself towards a large coordinated project. A tenured professor overseeing 15-20 students working on a project provides continuity that a student coding (who will find himself in a job in a few years, maybe even with friends or family taking up his free time) doesn't have.
University research projects are HARD to reconcile with the GPL. The BSD license (and similar licenses, like the MIT license) are great for universities. Universities producing money with corporate or government grants should produce truly "free" software, that are free for all to use as they see fit.
All Slashdot bull aside about keeping it free, BSD licensed code is FREE. Anyone can use it, nobody can lock it up. It's there, forever. If it is a university project, it will likely be always hosted, as opposed to small shops that can change direction/go under at any time.
GNU set out to create a Unix-like OS. The kernel got little attention, as the Linux kernel took off. Producing BIG applications isn't within the GNU project (and its programming offices at the MIT LCS (or AI lab, I forget where they are stuck).
End user applications require a different set of requirements.
The other problem with some of the tasks is that they are hard to divide the costs. Apache was spread among people that needed a web server, now the rest of us users leach off their work.
MS Office's costs are spread among all corporate America, with a juicy profit flowing to Microsoft. Creating a competing office suite is hard to manage, because spreading the costs is HARD. Licensing fees "spreads" the costs, but there is a free rider problem. Each Fortune 1000 would probably benefit by sending 10% of their Office licensing amount to an Open Office consortium as a long term 3-5 year investment in reducing their costs significantly, but each corporate board would prefer that every OTHER company send the money and they keep it.
Its the prisoner's dilemma, and its hard to fight.
Figure out a way to divide the costs, and you'll get quality GPL software.
Niche markets are easier, get all the vendors to line up for one solution. Mainstream markets are where "shrink wrap" software excels... there is a reason for that.
Alex
Right... because I can rebuild my Win2K desktops with my Compaq restart disk in 30 minutes, then do a network install of Office 2K. The developer can then install his favorite tools from the share drive in 30 minutes. Apply SP2 and hotfixes whenever they get around for it (before they leave for lunch).
If I was supporting a large Office, I'd have the automated installs. I got Windows to install off it (really hard, hit F12 on bootup, answer some questions, leave it alone) but got bored dealing with the documentation. Then for fun, update your drivers, I have them on the software share sorted by OS and model... I have 3 system models, all Compaqs. I ripped out the old Dell systems and put them to use as OpenBSD boxes long ago. Support is now a case.
Now, how much time did you spend installing new Themes under KDE/GNOME?
How often do you download the latest KDE/GNOME desktop to see how it works? How often is this spent compiling? How often do you decide to mess with the drivers and garbage to get sound working on Linux to support a $20 sound card (once had someone tweak for 2 days, instead of finding a supported card and whipping out the credit card).
When do you decide that you need to go "lock down" your workstation?
What happens when I need to show you the Visio diagram for a project? Reboot into Windows? Or did you spend 2 hours playing with VM Ware to get it working right?
Hell, I love my OS X workstation, but quite frankly, but Win2K laptop kicks ass as a client machine. I have network shares made available offline. This means that I can work on a client proposal on the train to/from work, do accounting work in Quickbooks, etc.
My Blackberry syncs in nicely, keeping my appointments and telling the server when it is hooked in.
I LOVE *nix. We build applications on OpenBSD/Linux. Both are great OSes. OpenBSD is the FASTEST server installation I've ever seen... My Penguin computing boxes are up running 2.9 in 30 minutes, then I just scp over the files that we've tweaked.
But you can't top the MS desktop for productivity applications. Sure the Mac GUI makes it counter intuitive, but I can deal. My software runs. All my hardware is supported.
I never lose someone for a few hours while they tweak their kernel. The only kernels that got compiled was the OpenBSD kernel. As I standardized that hardware, I scp it on to the new box and move on.
MS makes some good products. I stay off their server stuff because once I choose one, I have to choose them all. That's annoying. However, their desktops are more productive and faster. The WinXP test looks good too. The taskbar changes that they stole from OS X makes our virtual desktop program unnecessary.
If we can't access our files, we can (and have lost) $10,000-$30,000 projects.
I'll keep cutting MS checks. However, for a machine that NEEDs to be up, a Unix solution is the only way to go. Linux and OpenBSD seem to pair up nicely for us as a database and web server combo. They have great uptime, and no registry connections. However, for the desktops, I keep all the data on backed up servers, and if a machine dies, its rebuilt exactly how we want in a matter of hours. Meanwhile, with roaming profiles, we grab a spare machine while it is being rebuilt. It's a small price to pay to be in business.
Alex
Alex
Its really easy, you try to keep the execs out of the decision process. Get your requirements and develop a strategy.
Go buy the high-end Red Hat distribution. We did this for a client that knows we were using OpenBSD off a $30 CD. We also buy the nice Red Hat distribution.
Call up Red Hat and set up a support arrangement. They are a reasonable sized company. Alternatively, call IBM.
The advantage with Microsoft is that if you are a big company (say, the Fortune 1000 in the article) you get the special phone line for support, etc. Microsoft supports the Fortune 1000 a little differently than your pirated copies of Win95.
Don't sell them on free. Tell them that you worked with Red Hat's OS, and you found that it is better suited to this project. Inform them that you can reduce downtime (their real concern) instead of a couple grand on licenses.
More importantly, emphasize that it will save you time.
We don't use Linux or *BSD on desktops, it is too expensive.
Win2K or even WinXP involves a short installation procedure (before lunch) then come back and finish. Setting up a Linux desktop (for a technology guy, not end users) takes a few days of playing. Win2K tweaking with fun apps takes $1000 in software (including a $200 shareware budget) and you're good to go.
Look at your salaries. See what it costs your company to have you putzing around for days.
BTW: with the MS licenses and a point-and-click installer, how much time does it take to get another server up and running. Including your downloading Redhat over the corporate T1 (or whatever), how much of your time is spent putzing around on IRC, etc.
Sure, IRC is nice for REALLY hard problems. However, having a server down for 1-2 days while you troll USENET or IRC for help isn't acceptable.
Next time a MS solution is being proposed, try to get 24 hours to stall. Take the same list of software, and the budget, and CALL Red Hat Sales. Tell them what is going on, and ask them to put in a bid.
Alternative, call a Linux consultant, and work with them to put in a bid for the implementation AND for the Red Hat support contract. If the Red Hat fee is less, show that to management.
You all would get a LOT more credibility with management if:
A) You dress like professionals (I did NT Consulting for 4 years... we all wore a nice shirt and khakis... the Linux guys would often wear jeans, it makes a difference; my BSD shop does it too, it matters)
B) Emphasize solutions, not technology (they are looking for a solution, show that you understand this. Emphasize the savings in downtime, not licensing fees.
C) Focus on REAL cost savings. Don't CONSIDER unsupported downloaded applications. Discuss support agreements, Red Hat Network, etc.
Geeze, this isn't rocket science guys, understand what the executive is trying to accomplish.
Alex
Yeah, my bad, that is what I meant by the console market. Gaming is the first foot in, I would expect a follow-on product RSN (real soon now) that includes a large hard drive and TV tuner. It would be a UltimateTV/XBox combo with the DVD player. They might even bundle an MP3 player.
Once they have the system, all these other uses require no additional processor, etc. They can embed the equivalent of all these add-ons.
They need a strong gaming market to do this however. Why?
The processor/graphics are best subsidized by the gaming market.
Think about it, if there is a $300 price for the gaming, they can add the functionality for little more. They have the $300 Xbox, and several $400-$750 Xbox+ systems.
MS is always about bundling. The audio/video market is split into a low-end integrated solution and a high-end component solution.
MS will target the low-end (their specialty) with the integrated solution, and Xbox will be the first step. I would expect in 3 years Xbox2 which plays Xbox games as well as some new Xbox2 games. They'll speed up the process of console replacement, but game makers will just set whatever requirement they like. No problem that Xbox5 is out, some games will require Xbox1, Xbox2, Xbox4, etc.
The real trick for them, IMO, is getting the gaming market to make people pay for the processor. Once the processor/graphics is paid for, the extra features are just a matter of software and harddrive space, neither of which bumps up the product much.
Xbox is a console, designed to make money off games.
The long term goal is the monopolization of information into the home. Then they get a cut of EVERYTHING: games, music, video, time-shifting, etc.
As a whole-widget company, it will be increasingly difficult for single solution players to compete. Sure the high-end will never adopt the MS all-in-one solution, but Panasonic, AIWA, and everyone else that plays in the space is in trouble.
Expect Xbox based solutions to come in all forms... including those with a built-in amp to power 4, 5, or 6 speaker configurations. Some will include speakers, some won't. They'll create a family of solutions that share the same core and come in different bundles.
That's scary, if only because of their previous licensing strategy.
Alex
MS wants to sell the whole widget. Dell has profits. People buy Dells (generally) to run MS OSes. As a monopoly, MS is able to extract MOST of the profits. If they were a pure monopoly, they would get it all. MS has SOME competition, so they have to leave some profits to the OEMs.
XBox is a multi-billion dollar play to get their systems into homes. They are trying to monopolize the gaming market.
The problem is, without analyzing the console penetration, game developers won't know if Xbox sales are propped up by Slashdotters saying "damn the man" and buying them to play with.
If Xbox looks like it has great penetration (and 1 million "nerds" buying them to play with WILL be significant in the first 3 months), the games will come for the Xbox, and all us Gamecube owners will be left high-and-dry for third party apps.
Oh well, I've been reasonably happy with every Nintendo console because of the first party apps. My favorite was the NES followed by N64 followed by SNES, but I enjoyed them all.
The N64 was a failure, but I loved Mario 64, the first Zelda game, Goldeneye, Smash Brothers, and Hang Time. Those were enough games to keep me interested in the console.
I am getting a Gamecube for Blitz 2002, the new Smash Brothers, the new Hang Time (forget what this one is getting named), plus the new Mario and Zelda games. A new Metroid and Star Wars games are icing on the cake.
I haven't seen anything for Xbox that makes me want one. Halo looks nice, but I'll wait for the full featured PC version. I mean, anything good for the Xbox should be ported to the PC, and the PC will have games that the Xbox won't be good for (RPG/Strategy games that I really love). I have a HTPC for gaming in my system, so the TV isn't even an advantage for hooking up a console.
PS2 is starting to look good with some exclusive games. I may pick it up as a second system. The Gamecube just looks more impressive to me and has the games that I want.
Hackability? Give me a break. I want a gaming machine so my friends (the human kind, not on IRC) can come over and we can play a few games after work. You're right though, MS isn't going to be hurt by losses, not their style. When they have a high stock price, they use it to persue global domination. When they don't have a high enough price, they use cash to do so.
Their shareholders? Management controls enough of the shares to avoid a take-over.
Taking MS on head-on is likely a failure. Your best bet to hurt them is to create an alternative in either core or secondary markets. Buying a Gamecube helps fight MS taking over the console market. Buying Linux servers helps stop NT's spread. Keep MS contained, build alternatives.
The only reason to target MS is if you can find a way to stop them from coercing you. I'm not concerned that they make a lot of money. I'm concerned if they can dictate the Internet on the server side because of a client-side monopoly.
Alex
Microsoft has a monopoly on desktop OSes on the x86 platform. This results in games having to adapt to Microsoft APIs to compete.
Therefore, all PC games had to support the APIs. Microsoft is clearly leveraging its monopoly on the desktop to establish a monopoly in the console market.
Sure, the X-Box games won't pop into a PC, but you can probably share between 95% and 100% of the code between the versions.
I think that the dumping arguement is stronger, but the shared APIs help the case.
Alex
If the XBox truly costs $375 in marginal costs, and they are selling it for $300, they are engaging in dumping. It is illegal to sell products under marginal cost in an attempt to monopolize the marget.
A combination of Microsoft leveraging Windows (the DirectX API), a monopoly, into establishing a new monopoly through market dumping, should make this a no brainer.
MS got away with IE on a technicality. Because software, as we all know, costs 0 to copy, MS was able to give IE away without engaging in dumping because the marginal cost was 0.
They did spend bandwith. However, as bandwith is a fixed cost (you pay $X for Y Mb/s), the marginal cost for the copies remains zero.
I don't know if the idea of selling below marginal cost to make up backend revenues matters, but the entire Microsoft practice smells of dumping.
They are using their cash reserves and selling below cost in an attempt to put Nintendo out of business (Nintendo is a games company!) and cause Sony to bleed red ink.
This seems like a plain approach of market dumping.
The interesting thing is, most publically traded corporations are stereotyped at looking at the quarterly returns. Microsoft Corporation is unique in that it DOES look out for the big picture. The successfully leverage their monopolies and cash reserves to "cut off the air supply" of their competitors.
If Microsoft is selling under marginal costs to establish marketshare and monopolize the market (where they can then jack up the licensing fees, etc.) this should be clear cut.
Are any of the lawyers on Slashdot around? Am I on the right track?
Alex
Everytime a thread on Slashdot takes place about gaming consoles, 5-6 people get scores of 3-5 (insightful) for pointing out that game makes lose money on the hardware to sell the software.
This gets mentioned frequently, and 3DO is mentioned as an example of what happens when they don't do that.
I've always been wondering, do we really know that this is true? Has anyone any financial data to prove this?
If there is some proof, can the Slashdot editors include this information in each posting, so those of us browsing at higher thresholds don't see half the posts with this insightful fact?
I also don't really buy this theory. I mean, how much can the licensing fee be for each $50 game? Also, in an age of video game rentals, how many games does the average console owner own?
I mean, if you figure that the average game now sells for $50, the store pays AT MOST $35 for the game. The distributer probably picks it up for $25 (so the BIG stores get the bigger margins, no separate distributers). This leaves $25 to be split among the maker and the console. I can't imagine that the license is more than $5. Maybe it is $10? That would explain WHY Nintendo and (until recently) Sega made systems, $5-$10/game is a nice margin, plus they get the revenue for the author when they sell their own games.
Let's figure that the average console owner owns 10 games/console (that seems REALLY high BTW, I owned 30-50 NES games, but they were mostly the original $30 games, and their weren't rentals in the early NES days), plus rents enough games to result in the local store stocking an extra 10 games. This is 20 games/console, at $10/game, yielding $200 in licensing.
Now, how much of that licensing is Nintendo or Sony willing to spend subsidizing the hardware?
I had always heard that the stores make little margin on the systems (not a loss, but a trivial profit) and make their money on the games/peripherals. This makes more sense, as they trade a little bit of store space to get the margins on the games. The games are good for toy stores, as the space/product is minimal compared to real toys and the prices are high.
However, the console maker subsidzing the hardware (more than a trivial fee) seems absurd. I mean, MAYBE the launch versions get subsidized, but given the demand (preorders, unavailability for 2-3 months), why would they subsidize sales when they could clearly move the units at cost or above cost. Now, I could see subsidizing post-launch consoles to move sales, but manufacturing costs should go down over time, allowing the prices to drop (which they do) or the profits on consoles to increase.
Now, I COULD buy that the console makers sell the machines at cost. This would result in a subsidy of the "fixed" costs (R&D, setting up manufacturing process), but still, this wouldn't be real. As the costs go down (consoles stay on the market for 5 years, electronics go down tremendously in 5 years, but consoles rarely drop THAT significantly in price... i.e. a $300 console may drop to $200, but the manufacturing costs 2 years out should be half the initial costs), maybe they recover the initial subsidies?
I mean, the common Slashdot belief that the 3DO was $700 because of no subsidies and needing a profit, while the $200-$300 consoles are sold at a loss is ABSURD. That implies a $400-$500/console subsidy (or $200-$300 with an extra $200 in profit for 3DO makes), which would result in assinine losses.
In computers, processors are sold at a premium when new, but moved at lower prices after R&D is recovered and Intel was traditionally trying to move more units. This is simple price discrimination, not a value judgement on valid profits.
I mean, maybe you sell consoles at a loss to create a software market. Then the software market creates a demand for the console which allows profits on the later consoles, but this subsidized hardware "theory" makes no economic sense.
Realize that 3rd generation games are planned AFTER the launch (first generation hit with the product launch or soon after, the second generation normally starts in after the first round are finished before launch, and the third generation don't start planning until 6-12 months of sales are known), so creating a demand by moving consoles allows third generations products to be made.
However, I'd like someone to either provide EVIDENCE of this subsidy, or at a MINIMUM some economic analysis to show WHY companies would do so. This simple assertion ("remember, consoles are sold at a loss") is neither insightful or useful.
Alex
P.S. Apologies for singling out one of the posts, this could apply to any of dozens of Slashdot posts on the subject.
Right now, my DSL "modem" and Firewall/WAP sit in my study with one of the computer, and we use the wireless for the laptops. (I live in an 700 sq. ft. apartment, so for a home, YMMV) The plan is to move the DSL/Firewall into the entertainment center, which happens to be in the middle of the apartment because of the layout.
Originally, I was going to use 802.11b to get connectivity to the HTPC, etc., but with the growing availability of ethernet support in devices, I concluded that I should have a non 802.11b connection in the entertainment center. Their already is a phone line run back there to support the Replay, so splitting it and running the DSL there is trivial, and I will eventually have 3 devices in the system that need connectivity.
The Gamecube is going to support an ethernet addon, some sort of MP3 player may make it into my system (sure the HTPC - home theater pc - could do it, but it creates a UI nightmare... although having a Pronto makes life easier), and the HTPC obviously needs connectivity. I concluded that the USB-802.11b connection will be adequate for the docking station that's for the study, as well as the second computer.
Yeah, my freaky apartment has a docking station for my work computer, my fiancee's old PC that she uses for her Windows apps, my HTPC for computer gaming on the television, my laptop out of the docking station, and the fiancee's iBook. I guess it is a problem that I love technology for technology's sake, oh well.
The problem with 802.11b is that unless you have an intelligence antennae system, reception blows. My laptop whines at places that the iBook is fine at. Apple's engineering is impressive.
802.11b is amazing though, and more stuff should support it. I would love to get some of the LCD-based computers if I could get a good 802.11b setup, so I could display rotating digital photographs throughout the apartment. Oh well, one day.
Alex
No doubt, coding for IE is cheaper. As I posted in another thread, one of our clients made that decision, and now we have to rebuild the site to be Netscape friendly? Why? 10% of his "unique visitors" are Netscape, and they can't even use the site with the latest version.
If a 10% increase in profits > cost of implementing a Netscape version... well, Netscape version is coming...
Its a business decision. The IE5 version of the page is the low hanging fruit. Netscape is more of a challenge... Now if I could figure out the random Mozilla rendering problem...
Not a business problem, I'd just like to see it work under Mozilla/Netscape 6.x.
Alex
Come on, don't you remember all the OS/2 vs. NT 3.1 articles when NT 3.1 shipped? NT 3.1 was a flop, mostly used as a testing ground for people interested in keeping up with MS's new plans.
NT 3.51 was the first successful version of NT. NT 3.51 SP 5 was amazingly stable... it would be interesting to put an NT 3.51 SP 5 machine up against a Windows 2000 SP 2 (NT 5 SP 2) machine and compare.
Win32s was the backwards port of the core of the Win32 API to Win3.1. The two goals were:
1) Get new applications written against the Win32 API so NT (the future) would have some applications
2) Break OS/2 Windows compatibility layer... they kept changing Win32s until they broke OS/2, then they released apps for Win32s.
Windows 4.0 (Chicago AKA Windows 93 AKA Windows 95) was the version that combined DOS/Windows (to stop the DR-DOS onslaught) and introduced the Win32 API as the standard API. Win95 resulted in the Win32 apps that allowed NT to show some success on the desktop. NT 3.51 had some success as a server (very useful environment for managing Win3.1 desktops without the cost of Novell).
Win95 had some new APIs, which were mostly ported to NT 4 (except DirectX > 3 APIs). When I was at Citrix (MS Blocked WinFrame 2.0, then basically bought it to become Terminal Server), we couldn't support newer versions of IE because WinFrame 1.x was based upon NT 3.51, and IE required Win95/NT4 APIs.
Cairo was supposed to be the end of Windows with NT 4. Two years late and without a lot of functionality, NT 4 had (and still has!) some good server-side support and corporate desktop standing. When NT 4 lacked a lot of the functionality, MS declared that Cairo was a set of projects, not a release, and that some of them would be in NT 5. NT 5, two years late as Windows 2000, finally made a nearly API complete NT to match their home desktop dominance.
Windows XP appears to use a nearly identical system, focusing on a new user experience based on MacOS's improvements.
Microsoft has finaly achieved its 8 year goal of eliminating DOS support, ME was the end of the DOS based Windows, and it looks like all the old DOS games are finally dead. MS kept promissing better support for DOS apps/games in the next version of NT, but never delivered, instead stalling on their demise. Oh well.
Interestingly, NT 3.51 (I don't recall NT 3.5) was extremely portable, commercially supporting 4 processor families (this continued until NT 4, but the other platforms failled to take off).
The DOS support in NT, the NT VDM, emulated a 286, albeit much faster. This is the reason that you couldn't run fancy things in the DOS emulation, if it was a protected mode DOS API (386 DOS app), the NT VDM couldn't handle it.
Hopefully a better solution than VMWare (overkill, complexity, etc.) will exist to run old DOS games in emulation. My brother bought me the commercial version of Abuse (at one time a favorite) as a present, but I got it about 2 weeks after I migrated to NT 4 fulltime. Well, my new HTPC (home theater PC, just for gaming, I got me a progressive scan DVD player already) is going to be 98SE or ME based for gaming compatibility, so I guess I'll be able to play the old classics there.
Alex
This site is almost entirely a one-hit wonder. You hit the site, sign up, and leave on your way. We're trying to focus on the users hitting our site. The standard site is CSS/HTML compliant (some IE additions, but it should gracefully degrade).
We're looking into cleanup options to make it cleaner and therefore run on Mozilla, but Netscape needs a custom solution. Maybe I should get a WebTV system and try it out, we'll see.
Sending the user away from the page means that they aren't generating revenue for my client. We're not interested in improving the web, we're interested in improving their bottom line.
If it works in my Mozilla browser, terrific, if not, oh well. If and when Mozilla/Netscape 6.x provide enough of a reason to make the site compliant, we'll work through their bugs.
It's annoying, but IE/Netscape 6 conversions should be easier. I don't mind (too much) writing two stylesheets. They don't take that long. It's making two versions of the site (a legacy one for Netscape) that is annoying me.
I test in IE because thats what the users are using. I'll develop for Netscape 6 when the platform is available.
The central codebase is the same, I just need to write different HTML renderers...
Sigh, one of our projects is to write our own XML language that was a content/display combo that wasn't HTML. Then we'll just write three renderers, IE/Netscape/Mozilla. Oh well, one day.
Alex
I spent months trying to understand the tech before making a purchase. I then spent months trying to understand how to best use it after I made a purchase.
:) Hell, anyone, regardless of background, should be able to find information if they dig deep enough.
I appreciate that I don't fully understand it, but I won't apologize. You guys that work with the technology would be really helpful if you provided information (like this post did) instead of just telling us that we are wrong.
Sit on some of the home theatre boards. The HDTV engineers on them disdain people that don't understand things as well.
It shouldn't require 4 years as an engineer at an HDTV firm to understand what you are buying... My four year MIT EECS degree should be fine...
I find your arguement that interlaced is more pleasing fascinating. However, I would still rather the information be transmitted at 480p. 480p is twice the data as 480i.
The reason that I find 1080i interesting is that it is quoted as an interlaced spec. Many people think that 1080i is better than 720p for this reason. This is blatantly false. A 720p image has more data.
Ideally, you want to get as much data as possible into your system. Once it is in your system, you want to be able to customize how it displays based upon your preferences...
For example, I may want everything to display at NTSC levels through my VCR, but I still want the most data reaching my house. I (through my equipment) should be able to do what I want to display it best, but I still want the most data in.
My output is at best what I get in. GIGO...
Alex
One of my client's sites was written with just IE in mind. It makes heavy use of CSS, and Netscape's CSS bugs just cough on it.
.5% and 1.5% of this website? They probably aren't worth spending resources on beyond testing on the Mac, but you have to evaluate your costs.
However, the logs indicate that currently 8.5% of our users are Netscape 4.x.
The operations guy at the client broke out his calculator, saw the costs of my fixing the system for Netscape, saw the revenue/profit increase, and saw that B>A and said, do it.
I was hoping to just change the style sheet, but Netscape is totally busted, so it looks like separate scripts. Sure the IE version will be the priority, but when you can increase profits 8-10% of more (in fact, increasing revenue by 8% should increase profits 10%-12% based upon some fixed costs, etc.) it becomes really hard to justify ignoring.
Unless technology costs are a rediculously high percentage of your budget, you can't ignore 8% of the market.
Now WebTV and Mac, that are
What about non-commercial sites? Code to HTML standards, and use minimal CSS. While we have sites that need heavy CSS to look amazing, the site could work without them. Limit yourself to fonts, sizes, etc., and you'll be fine. Don't worry about it looking right tot he pixel and you'll be fine on multiple browsers.
Alex
Okay, a standard television can do 640x480 interlaced (called 480i if you follow DTV specs). This is 60 fps interlaced, which is really 30 fps. Alternatively, you can do a 240p signal in the 480i system, which is what the original Nintendo system did. (IIRC)
:)
With a HDTV-ready system and a Gamecube, you can do 480p, which is 640x480 progressive, which can provide a true 60 fps signal.
Because of the size and shape of my living room, I have a standard size television, which is a 4:3 ratio. Some of the HDTVs are full-screen (4:3) and some are wide-screen (16:9). You can send a 16:9 signal to a 4:3 television and it letterboxes (or you swap the aspect ratio and get tall, thin people). You can send a 4:3 signal to a widescreen and reverse letterbox (on the sides) or stretch it and get short, fat people.
In supporting 480p, the Gamecube offers a true 60 fps, as the entire screen updates every frame. With a standard television, running at 480i, you will really only effectively get 30fps, as it takes 2 frames to draw a full image.
Interesting, the 1080i ration for HDTV (which can also carry a 540p signal) is interlaced. Interlaced is fine for things without that much movement, games and heavy movement systems benefit from progressive images. Computers have all but abandonned interlaced signals, and with HDTV, you'll have to find a TV/STB combo that does what you want. The 720p resolution will be amazing for HDTV and the "next" generation consoles after this crop.
Alex
In high school, I used to check Tom's Hardware Guide regularly. His posting on over clocking seemed fascinating. He seemed to have a good understanding of hardware issues.
Then I went to college, still a reader.
Taking a few courses on hardware however, made reading his reviews painful.
He doesn't understand hardware, and it shows. Sure, he is able to run tests, but his reasons are completely flawed.
I don't accuse Tom of being on the take... However, I think that Tom should stick to testing, and not give his "engineering" insights that are based upon made up terminology and without an engineering basis.
This is a more reasonable review, and the first one that I viewed of his in years.
Alex
I've been evaluating Mac OS X for a few months. Our applications are primarily web based, though we do Java and are adding Qt to the technology base.
We had each programmer with a Linux box and Win2K box. Now it is hit or miss. If you need/want a dedicated server, its yours. Otherwise, you use a development server.
Mac OS X offers some interesting possibilities. One system, with a full BSD subsystem (adequate for us), as well as an Outlook Client and MS Office (coming to native OS X, "Real Soon Now"). When those are released, we're really going to consider Powerbooks instead of the Compaq Armada's with docking stations.
Giving everyone a full Unix desktop (complete with CVS client) as well as Office Productivity apps would be great.
VirtualPC would probably even be sufficient for the few Windows only apps that we need. (Test the site on IE/Windows, Quickbooks for my accounting, etc.).
I realize that if you are doing real Unix coding, you need the Sparcs. But for those of us that just need a Unix-like environment, it is a sufficient development solution.
Alex
Despite deploying and building our systems on UNIX platforms utilizing BSD/GPL systems, we have an NT 4.0 network with Exchange 5.5 running.
Why?
We need the coordination ability of Outlook. I haven't seen anything for UNIX that compares. The ability to schedule meetings, observe schedules, and otherwise coordinate things is extremely useful. I would love to replace the Exchange Server and backend infrastructure with Free Solutions, but they aren't there yet.
I use IE and Mozilla as my browsers (IE 6 is pretty flaky), but I am stuck on a Windows 2000 desktop (albeit with 3-4 SSH sessions going) because of collaboration tools, accounting tools, and office tools.
This is a step in the right direction for those looking to build Free solutions.
Alex
Jabber, Inc. (jabber.com) is certainly an obvious choice. Keep in mind that the RFP will include what you need, not the technology. Any company that works with (supporting) Jabber is a good choice.
Anyone can take the Jabber source, modify it (if needed), and install it for you. While Slashdot would probably like you to go to the authors, in looking for your corporate bottom line, you should contact ALL companies that work with Jabber.
Sure the source is "free", but who cares?
What matters in this case isn't the license (though the GPL may make the Jabber-based bids more pleasant), its the result.
Each company should submit you a proposal with a description of how they will meet your needs and a price.
All the companies that submit with Jabber have an advantage, they don't need to support the full infrastructure of development, just their portion. The Jabber, Inc. people should be able to bid competitively, given that they already have a sales and support network in place for selling Jabber.
Good luck, and thank you for considering open source solutions in your RFP process.
Alex