Re:The correct pricing structure for most software
on
Pricing a Software Product
·
· Score: 2, Interesting
Well, not to pick nits, but "By giving away software for free you are reducing one of the barriers to entry in your market.", is completely wrong. You are lowering the barrier to adoption of your product (which is good). You are also increasing the competition in the market. However, the "barrier to entry" is still the same (it'll cost the same if you do the same things weather you sell the product or give the product away), which is the largest portion of barrier to entry.
Some one really ought to tell Microsoft that they can't complete with similar products by charging money. Last I checked, MS Office has probably 2 or 3 legitimate competitors that are completely free. MS Office rakes in cash like very few other products.
While there are plenty of other ways to make money, generally for niche products, charging directly for the software is a good idea.
I think Cygwin will work just fine. I've known a number of people who used it for extended periods of time. It'd be more helpful to know precisely what it is you are planning on doing to know for sure if it will work.
However, in a larger context:
Uhhh, you are taking on a customer for whom you have no tools and no infrastructure for? Who doesn't fit your current model, and fundamentally doesn't fit how you do business? Unless you are laying the ground work to bring in lots more revenue at a lower cost in the future, this might be stupid to do.
Now, a company has to grow, but remember the princepal that says, "Not all customers are profitable". You don't want customers who don't make you money. I remember a story about an advertising company that eliminated 70% of their existing customers and have revenue plumet, but their profits jumped by 30% (as a dollar value, not as a percentage of revenue, they made 2.5Mil instead of 1.5Mil in profit, I believe revenue went from 30Mil to 12Mil).
I know on more then on occasion, the smartest thing the guys in charge where I work is to fire customers. Some customers aren't worth the time or the trouble to deliver service to.
This isn't an anti-Window post, it's merely a matter of considering weather or not this is an area you are planning to expand into, or if this is a one-off, non-scalable solution for a single customer just to get the business.
We run into this quite often, around it's driven by sales people whose sales goals are about bringing in revenue, not bringing in profit. If it costs us $1000 in to bring in $500 in revenue, that's a stupid business proposition. If it's a big chunk of revenue, and you can build it while making money go for it.
Uhhh, you might want to go search those links a little closer. The first two don't have the word "icon" in them according to Mozilla's "Find" feature. If you're really talking about "Icons" (the crux of the Point and click concept).
The third one does, and but only in the context of dragging icons and double clicking them (in the June of 1981 line item). It comes up tangentally later in the 1988 and 1991 sections amount Microsoft. That particular line item in June of '81 I believe is referencing a computer made by Xerox not Apple.
Any chance you'll point out the specifics of the text that clarify that the Mac's didn't specifically take the concept of Icon's from Xerox? Or that Job's inspiration for developing a GUI based computer didn't come directly from his visits to PARC.
Heck, even Jobs openly admits that at PARC, they showed him three things, and he was so blinded by the GUI that he didn't even notice the other two (OO programing, and networking).
Search for the text "three things". It's right there. Now, Raskin did work there for a little under two years before the PARC visit, but unless it's the "PITS" thing, I don't see anything that leads me to believe that Apple didn't get the idea of a GUI directly from PARC. Raskin might have had the concept in a design 15 years early in his Ph.D, but PARC appears to be the one who shocked Jobs into realizing it was a revolutionary idea. So in the end, it appears PARC deserves a lot of credit you seem to want to deny them.
First, off, I'm not a finanical instruments experts, so I can't say authoratatively why they aren't using a bond. I have several guesses.
Bonds give you no authority (in Google's case, that's relatively true anyway). Bonds are also a lot less likely to make any big money for the investors. They'd be a lot better off to privately finance the thing thru a bank. A bond is nothing more then a loan with terms set in a bond where the debt is something you can sell.
There's in theory, low risk, and low reward. It's a different type of investor. An IPO for the company has absolutely no risk. They give up shares (paper) of the company, and they have no liability. Google probably doesn't have the assessts to back up a bond (their value is in their algorithms and the people who run their systems and some data that will be out of date 6 months from the date of purchase, not in physical assests that can be sold during a bankrupty fire sale).
The reason to IPO is to generate income. That's why the company does it. There are other reasons why. It now allows the owners of the company to selectively cash out to a liquid market. So the original investors can get in and out pretty easily. If they didn't want to generate money, I'm going to imagine, they could just get listed as opposed to actually issuing new stock to be sold. Then the original investors could get in and out as they pleased. They could have a much smaller IPO. You do it go generate revenue.
In theory, you do actually have voting rights with Google, it's just that they don't do you any good. At some point, that could actually change, and you still actually get to vote. I'm guessing if every non-founder stock holder votes one way, the founders might consider it.
Microsoft had no dividend for 25 years. I'd be just incredibly happy had I bought them on their IPO. As a long term investment, Google might have similar possiblities. Google I would imagine is going to start accumulating incredibly valuable assests either by creating them, or buying them up.
I don't think that Google will go up in price. They used the same system that the US Gov't uses for selling bonds in order to virtually guarantee a solid price. However, speculative stock buyers along with pre-IPO shareholders who slowly dole out shares while the price runs up. They'd prefer to see Google sell at a $10 price point. There is a lot more run up room. Once the stock moves from 10 to 100, how many people will want in on it? It'll be a feeding frenzy, back like it's 1999.
The person I was responding to, wanted to know why they should care which way it goes (Dutch Auction, or standard IPO). That's why they should care. One way the stock they purchased actually contributed to the value of the company, the other, it contributes to the pockets of a bunch of bankers. If you can get in on it the first day via normal IPO, you'd much prefer that Google use a standard IPO setup (there's plenty of money to be made if you can get in on the low end). If you are a long term investor I know I'd rather have my company get the $3 billion, rather then the investment firm and their friends. I don't own any shares in the investment firm or their friends.
I think Google is silly company to invest in, given that I don't get much in the way of voting rights, and they aren't planning on having a dividend. There's also so serious upside. They could be the next Microsoft. I know I'd be just thrilled if I'd bought every last share of Microsoft I could afford in 1990 and held on for dear life during the ride. It'd be worth about 1000 times what I paid.
Google practically mints cash. They have incredible technology that no one can duplicate. They are insanely popular, and have cornered their market. The problem is that switching search engines is trivial. If someone else can out-google Google, it'll be like when the car was invented. They'll be the best damn buggy whip maker their ever was....
When you buy shares of Google, you'd really like Google to get that capital. When you purchase shares of google, you are now an owner in google. It's now in your best interest to be sure that google win's the tug of war between who gets the money. Because it'll maximize google's value.
This isn't so true if you're a speculative buyer who things that Google's price is going to jump up, and if you can just get your hands on it, to turn it over days later while it's on the way up. Then your on the wall street side, and you'd like to see them win.
So yes, depending on the type of investor you are, you have a vested interest in seeing one of the two of them win.
Hopefully, the price won't be the result of playing the games with supply and demand, and the psychology game that happens on Wall Street. There shouldn't be a sky-rocketing value, that if you can get your hands on it, in the first 3 days, and sell hours later a huge profit can be turned.
Liquid markets with stable pricing is good for everyone in the long term. Wall Street's problem is that if your plan isn't going to make money for Wall Street in the short term, they aren't interested. Short sightedness will be the financial ruin of this country if we continue to do things to maximize value in the short run to the detriminte of the value in the long run.
You're doing a very bad job of interepreting his words, so I'll give you an example:
1. Item #1 2. Item #2 3. Item #3 4. Item #4
Now, that would look just stupid if it was like this:
1. Item #1 2. Item #2 3. Item #3 4. Item #4
In the end, ASCII (which all good e-mail should be presuming it's in English), there is no such thing as a hard linebreak versus a soft linebreak. In desktop publishing such a thing exists. E-mail as a general rule doesn't have a well defined, broadly implemented standard for such a beast.
Which is precisely his beef. It's relatively straightforward a paragraph all on one line into arbitrary length lines (in this case, there are only hard-breaks, and the text viewer implements softbreaks for you). However, it's painful to join things that aren't 80 characters together. The text viewer can't tell if you meant that carriage return to be a softbreak or a hardbreak. On a hardbreak it shouldn't join the two lines, on a softbreak it should. A lot of people aren't terribly consistant (I'm not for example), and their e-mails are a pain to read in an auto-formatted text viewer. I've cringed on more then on occasion when someone quotes my e-mail on a list when they use a 60 char line breaking, and I use 72 char line breaking. Then the last 12-20 characters get split onto their own line. It just looks horrible.
There is some header that can affect this along lines of "Text-flowed" in the Mime Headers, so that things word wrap properly (essentially, your text has soft line breaks, and hard line breaks), where as standard ASCII has no such concept.
I use mutt with vim as my text viewer. I've been known to re-format e-mails on a fairly regular basis and re-save them to be readable to me. I do the interpretation of soft vs. hard breaks and save it so it's easy to read (for me).
Stop being so hard headed and assuming the other guy's just an anal retentive prick. He's got a legitimate point that the interpretation of line breaks is very difficult.
It's simple. You can't sell your vote, if you can't prove the way you voted to someone else.
In the olden days, people would sell their vote for money. It wasn't until I believe the 1850's or 1860's that we had an anonymous voting system. In an odd coincidence, we imported the Austrialian method back then too!
Before the 1860's you wrote in the name of the candidate you wished to vote for. In small enough precinects, you could literally know everyones handwritting. Before that, you actually walked into the town capital building, and announced your vote in a loud clear voice the the people in charge of keeping track.
Each candidate would have a witness there keeping track of who voted which way, and could then pay off the people who they bought a vote from.
As the other response said, I'd imagine that the first whites to vote for a black in Georgia probably didn't make it too far out of the voting booth before getting harrassed. Unless there was an anonymous system.
Sure a Gigabit card could completely saturate a PCI bus. I'm well aware that's why they are built into the northbridge.
However, generally when benchmarking one doesn't actually use both the Gbit NIC's (he's a double idiot if he failed to mention this). From the setup, he's got something silly going on. As a general rule, the NIC and the harddrive will need to use up roughly the same about of PCI bandwidth (they are writting roughly the same amount of data, plus or minus framing/headers for the frames/sectors). 140Mbit/s * 2 = 280Mbit per second. So your telling me the "other misc" peripherals are using up 2-3 times the PCI bandwidth of a Gbit card and a harddrive running flat out? What else do you have hooked up to your machine?
More likely, he's running out of CPU, interrupts (they are queuing up faster then they can be serviced), or scalability in the kernel somewhere else, long, long before he's running out of PCI bandwidth. You took the two biggest PCI bandwidth hogs on his machine and showed they used up possibly a fourth, (maybe even a third if you throw in lots of overhead) of the PCI bandwidth, and then claim the "extra stuff", takes up the rest. I'm not buying it.
I'd hear what your saying, if it wasn't just silly. You can't even get close to the Gigabit thru put if you don't have a 800Mhz machine (he doesn't say how fast his machines are). I'm not sure if that uses up all the CPU power of the machine or not.
I've had exactly the problem he describes over NFS with 2.4Ghz Xeon's machines with Gigabit cards on the Northbridge. In the end, it was that NFS protocol has latency in it. In Ethernet parlance, the sliding window is too small. So while I could easily in aggregate get a Gigabit of bandwidth in a number of connections, I could never get a gigabit of bandwidth in a single connection. I wouldn't be shocked if this is also the case with Samba.
I could put about 500Mbit/sec over the cards via NPtcp, but for a standard file transfer over NFS, I couldn't exceed about 14-15Mbyte/sec, which is ~ 120Mbit/sec. You want to get the harddrives, and the PCI contention out of the loop to see what the problem is. Run NPtcp on both ends, check what the maximum thru put is. That'll give you a good number to start with to see how much data your machine can actually push over the wire. From there, it's easier to start creating artificial contention and CPU usage to see what could be causing the problems. If you believe it to be the PCI bus, you can fiddle with the timings of the PCI bus to enhance the bus utilization. Read up on it here:
http://www-106.ibm.com/developerworks/linux/librar y/l-hw2.html
I have several harddrives that top out around 14-20Megabytes per second, which turns into roughly the speed rating you are talking about.
I doubt your running out of PCI bandwidth.
It could be the latency, or that you have a poorly tuned network stack. I know that using NFS, getting 12-15Mbit/sec was considered pretty good given the inherient latency of the protocol.
I had similar problems no matter what protocol I was using FTP, HTTP, or scp. What I found was that, I needed to use a network speed tool: NPtcp, which is part of the netpipe tool set.
The other thing is to figure out if your cards support Jumbo frames. If they do, it can be a boon to go change your MTU, and modify specific parameters in your TCP stack, and in your application to change the socket options used (specifically, to use a packet size larger then 8k). I'm not sure how to do this under windows, but I've found it readily available under Linux on google searches.
More information is more useful. Knowing what chipset it is based off of, which drivers you are using, what OS, would be mighty helpful to helping you solve your problem.
Uhhh, I'm not going to argue that it's bad for work to be oursourced. I know enough economics to understand, that fighting that machine is a losing problem. You'll be crushed one way or another. You might as well embrace it, and deal with it as best you can.
Look how well it worked out for the Autoworkers Unions from the 70's and 80's. They were a powerful group that got absolutely crushed because they thought fighting a free market economy was a good idea.
Now, OpenSource work is good for the economy, if not good for the programmer. It takes highly specialized software and turns it into a commodity. Take a look at it. You can now purchase a computer and load it up with OpenSource software enough to make people highly productive for less then $300 a computer. Take a look 10 years ago. You couldn't hardly get a computer for under $1,000. That's a huge amount of capital that is no longer needed to run a business. Which means the business should have higher profit margins. It lowers the barriers of entry into a marketplace. Lowering the barriers of entry, increases the competition, which forces an increase in the efficiency. That's very good. You want a more efficient economy. It means lower prices for goods, and more capital to start other new ventures with.
Open Source is good for the economy, if not for the programmers. Open Source is good for the economy, in the same sense that roads and bridges are good for the economy. Resources that are generally helpful to everyone, can be built once and only once. That's wat open source is all about. A stable operating system is a necessary tool for almost any office job. A word processer, a web browser, and the myriad of other tools that are broadly used act as a tax on business if you have to pay for them. Which is good for you, if you're the tax collector. It's bad for everyone else. So in the broadest sense, Open Source is bad for programmers, but good for the rest of the world.
Besides which, there will always be a need for customized software which works in specific ways. That process will probably always need highly skilled people who can either write new modules, or extend and configure existing modules. It's sorta like saying, well the car's going to put the horse carriage drivers out of business. Yeah, but they'll be taxi driver jobs coming. It's a cycle, be prepared to adapt and change. Make a lot of money, save a lot of money. Just like, we'd need a lot more programmers to get stuff done if we wrote in assembly. You don't hear anyone clamouring to pass legislation to force people to write in a assembly. Next, you'll be telling me, that John Deer tractors should be outlawed. It's all about efficency. A word processor shouldn't cost money. Society should foot the bill once and make it a public resource. Just like public roads are.
Googling around, that's the best link I've seen, but it's relatively short. I haven't looked into in about a year. There is a patch to mount that will allow you to use let you setup cryptoloop (the loopback block device with encryption built in). The maintainer of mount (Andries Bouwer), however wouldn't accept the patch for some legalistic reasons (copyright and/or encryption, I've forgotten the details). Try "man losetup" for information on how to get crypto loop working.
What you do, is have no swap partition, then at the end of the bootup sequence run losetup to setup loopback encryption to either a file or a partition. Then initialize the swap file or partition using mkswap, then swapon/dev/loop0 (or whatever loop device you used).
So something like this:
losetup -e DES/dev/loop0/tmp/swapfile
mkswap/dev/loop0
swapon/dev/loop0
However, I can't seem to get that working, as my RHEL doesn't have DES as a modules, and losetup doesn't support AES yet. So there's probably a kernel patch to track down. However, it sounds like Mandrake supports this out of the box.
Finally, I'm not much of a 2.6 user, what I described above will work in 2.4. 2.6 I believe either has in the main tree, or it's coming "Real Soon Now", a feature called dm-crypt. dm stands for device mapper. It's the low level replacement for the guts of LVM from the 2.4 kernel. It sounds like a layered/plugin type feature, one of which you can put LVM on top of. Some clever fellows decide to put crypto on top of it. I don't know a thing about that, and have never used it. However, search for dm-crypto in google, here's the most useful page I found in a single query:
I've always like the "redhat-config" series of tools on RedHat or RedHat derived products.
chkconfig and service are valuable for doing things.
yum, apt, autorpm are all nice tools.
logwatch is great.
PAM isn't an application, but using PAM and LDAP can make having users spread across machines is a snap.
tripwire is a great tool
kickstart does wonders if you have to install lots of machines. I use it to completely document every scriptable part of an install.
WebMin is great, but it worries me from a security perspective.
gq is a great LDAP editor that I use to edit LDAP entries for users.
Software like Bastielle Linux (a script that attempts to harden a machine)
iptables, iproute2 (including ip and tc) are wonderful for networking.
cron, sh, sed, awk, perl and python are used in conjunction quite a bit.
fuser and lsof are used frequently by to to figure out what is going on.
ethereal and tcpdump are tools of the gods.
ssh is a thing of beauty.
encrypted swap is fun.
Nagios, MRTG, and sar are very useful for profiling and monitoring of your machines.
Who exactly is your target audience? People who've never seen UNIX, people who've worked on UNIX environment for years? What is it you are attempting to accomplish with them?
Most of my list would be boring to people who know a lot about UNIX, however some of them are Linux specific.
As I recall, in C x[y] and y[x] are defined to be identical. I believe you can do 1[y] and have it work. That's because a[b] is must be identical to *((a) + (b)). I'm over using paranthesis intentional. Because addition of a pointer and a constant is communitive, either one works. Because (1 + y) is a legal, and returns a pointer it works.
Personally, I think it's a completely crappy thing. You should get an error telling you: "Attempting to use a constant like a pointer".
So it should print out "10\n10\n";
If that's an ANSI C complier: you should get a warning about no return in main, an illegal declaration of main, and an unknown function printf. You might get away with main, because I believe all functions implicitly return an int.
Yes, but you're probably still breaking the employment agreement you signed if you work in the US as a programmer.
The company owns the copyright on the code your write for them under the legal concept of "works for hire". Unless they failed to pay you for it, they own it. You leaving the building, and letting someone else use it is a copyright violation. The employer you originally wrote it for should be able to successfully sue for copyright violation.
Now, I have plenty of ideas stuck in my head that I've ended up re-implementing in nearly exactly the same way at two different places. I can re-implement my C++ database independent access library in a little under two weeks. I can write the reference counting classes I like to use in a day. I can implement the interface wrapper classes in about 3-4 hours. I can write the socket stream libraries, and the various SSL varients in a day or two. I can write a simple logging scheme in about a day or two. They are written at different times. I can show them as they are constructed, and demonstrate that they are clean room implementations. So I should be in the clear according to what I've read and been advised about the law.
I ended up using the same names, and structuring the code the same way. Some one pointed out I might be breaking the NDA I signed at a former employer. So I ended up re-implementing it all again, this time going out and finding the functionality/API I wanted in publically accessable code (and implementing it from scratch to avoid licensing issues). I documented where each API/functionality idea came from, they we're strewn across several programs. This also showed that the ideas represented "current well known techniques", which no NDA can cover.
You can use exactly the same ideas, and you can implement them at home, and take them into work and let them use them (put in the copyright that you as an individual are the copyright owner, and they are fully licensed to do anything and everything they want with it in royalty free). At least then you'd be legally doing it.
You might also want to try tigris.org. They don't appear to be officially supporting tigris, but I know that a bunch of people there are involved pretty tightly with Apache.org, and they are a "development shop", that supports several products. I'm reasonable sure if they won't do it, they'd know who would (the same with Apache).
First off, what difference does it make how many IP's it is? Move 10 DNS names, or moving 100,000 DNS names is pretty straight forward to do via "sed" if you use BIND or most any other text based DNS system. Even if you don't use one, you can change to one by having it be your backup, then use a text editting tool, then switch the roles of primary versus secondary.
Second, the company and the court are forcing this ISP to violate other contracts they have with ARIN.
Third, why doesn't the company have a migration plan? Why can't they take care of it themselves? Migrating services and servers is a pain, but can be done relatively transparently. It's expensive, but I'm not sure how suing your vendor and asking him to pull off nearly impossible technical feats makes this any better.
I have no idea why the ISP has to be complicit in the DNS changes. If they are a hosting company, they should have all of the technical expertise to manage this change without any help from the ISP. All they have to do is continue to pay the original ISP, until they can move to a new ISP.
Finally, any sane person would probably have used mapped IP's so that internally you would have just moved the network and change the mappings at the edges of the network.
The webhosting company is going to become a total screw job. If they had any technical sense, they'd realize that even if the company complies with the requirements and lets the block of addresses be changed, it's still going to be a serious problem with the BGP routing unless they are moving a fairly large block of IP's. If they are moving a large block of IP's, why didn't they just apply for their own? Why don't they apply for their own, and migrate to them in-place? Why is it the ISP's problem to arrange for the technical problems of the client?
Please stop referring to spam filter comparisons as "science".
I believe the author of the article would have two issues with that assertion.
First off, you can have science about how fast grass grows. You have science about how many sexual partners a person has. You have science about how to manipulate people with irrational arguments. Science can be applied to anything that you apply scientific princepals to. Science in a lot of ways, is merely a matter of measuring in a controlled manner and then commenting on such measuring. The usefulness of science is when those measurements are useful and applicable to common every day situations. Like say, your twice as likely to die in a car accident at 50MPH, then 40MPH.
Second, the author sounds like a mathematician, and somewhat of a scientist, and he has a mathematical interest in the filtering of SPAM. It's just as mathematical as using markov chains to model queuing problems to measure how long you'll have to stand in line at the checkout counter. To him, it's an interesting mathematical problem, which in a lot of ways, means that for him personally SPAM classification and the comparison of SPAM classification techniques IS science.
Finally, the results the author is referring to, are due to be published in a peer reviewed journal if I understood it correctly. So in a very technical sense, it is in fact being published for scientific review.
I think a lot of his issue is that you can't use the results of that paper to draw any useful conclusions for yourself if you aren't in a similar situation. As an example, I can get about 18 gallons to the mile in my F150, even though it's only rated for 13/15 city/highway. I manage that by setting the cruise control at the speed right after I switch into the 5th gear, turning off the A/C, driving on predominately flat roads, buying the highest rated fuel, and not stopping for any reason other then purchasing gas. So I could publish a paper saying that a F150 can easily get 18 miles to the gallon. However, that's incredibly useless to anyone who doesn't realize the conditions they have to drive in. His argument is that, the paper doesn't represent the results anyone else would get.
I don't think you are correct. If I can control the POST sequence, and I have the Microsoft Software, the system can be broken. Period.
It's the ability to flash the BIOS that will make it happen. At some point, Microsoft will have to trust a piece of hardware. If they trust the software, it's merely a matter of time to find out where the branch is that says "yes this is trustworthy", and change the binary so that branch always takes "trustworthy" choice. Just like if I have access to your GPG binary, I can say that a message I sent you is in fact signed by Microsoft (the element of trust everone forgets is that you have to trust the binary sources, in this case, Microsoft can't, as I can fiddle with them). This is an arms race that Microsoft will always lose, it's just a fact of life.
So they must trust a piece of hardware at some point. That hardware must be untamperable, with no way for me to interject myself between it and the Microsoft hardware. As soon as I can interject myself between Microsoft and that piece of hardware, I've won. If I have access to the BIOS, all I have to do is setup some type of virtualization software (Think VMware). At this point, all I have to do is emulate the piece of hardware, and jigger it to always say: "Trustworthy" (essentially a MITM).
If you don't believe that type of attack is plausible, then remember also, I control the client, at some point, I can attack the PKI system. I have access to the PKI portion. At some point, you must have absolute trust of the PKI system, I have the client, what would it take to beat that system? Does Microsoft keep it's list of keys someplace around (it has to, I can subvert that)? That's like giving me access to the root cert's for your Web Browser. You'll trust my hacker sites if I can insert my key into your list of "trustworthy certs". At some point, if I have access to the boot sequence, I can break the system.
The only way it could be secure is to have the hardware have the list of trustworthy keys and have the hardware never give up control to anything that is considered untrusted.
How does Microsoft check that they are running on such a trusted? At some point, they either have to trust the hardware implicitly (which I can fake), or they have it in software that I can modify. At that point, it's either making an untrustworthy piece of hardware (or emulating one), or fiddling the bits of the software. In the end, DRM is a losing proposition. All DRM systems will be broken.
Microsoft might be able to encrypt the software, and only allow it to be decrypted by modules hardware that has the public key embedded inside. However, somebody will just tear the thing apart, or use an X-ray machine to just extract the public key (which at this point is merely a secret piece of data, not really a public key).
Actually, go look at it. I'm eligble for "A plan" from Ford (essentially, I get the best deal Ford offers to rank and file employees). My brother-in-law works in Ford Motor Credit, and can extend that plan to "Friends and Family".
I bought a truck essentially for the reasons you describe, and now I'm not sure it was the best idea. However, leasing was a very, very good deal for the first 4 years I did. Ford encouraged you to lease heavily. I think the buy vs. lease payments for my first truck was something like: $315 vs $500. Ford priced leasing so attractively, that it was very hard to turn down. They had rigged the system, so that leasing was always a winning proposition (I never paid for full depreciation value of the truck while I had it). Ford was always losing money on it, and hoped to make it up on the used market. The problem, was nobody wanted to pay the inflated used price (to make up for the value I didn't pay for), because it was a cheaper payment to lease then it was to buy.
The buyout price on my first truck was $450/month for 4 years. So I always got more value then I paid for, but theoretically, I wasn't getting the best deal I could have. However, I had a fixed price, full warrantee, other then gas and oil changes, I had no other costs. About the only thing that sucked about it, was you get screwed on miles. You either pay too much or pay too little. On my first lease, I didn't get enough, and ended up owing an extra payment for it. On my second lease, I ended up with an extra 12K miles I paid for, but never used. However, if you are a buy a new vehicle every 6-8 years kind of person, and don't put that many miles on it, buying was just stupid. Over the 6-8 year time frame buying was more expensive then leasing, even over the long term. You have to plan on owning a vehicle for 8-10 years for it to be a money saving proposition.
Leasing was a great deal 4-6 years ago. It was priced too cheaply to not do. At the time, not being able to drive to work was a serious problem. The extra reliability was worth it to me.
Now I'm trying to save money over the long haul, and I live within walking distance of work. So I bought.
Uhhh, you're a complete idiot. You know this right? Without knowing how many shares available, and what percentage are going to be available to the public, there's no way you could guess what the price of the shares are intelligently.
See, if Google released but a single share, that represented 50% of the company, I bet that single share would be measured in billions of dollars.
If they release 100 Trillion shares, my guess is, fractions of a penny will be the value.
Now it is a good idea to keep your stock price in the $5-$25 price point as it's then a pretty liquid stock, because most investors can afford lots of 100 (generally the smallest unit stocks are sold in by brokers, breaking a lot costs you extra). Institutional investors like pretty liquid stock prices, as they can get in and out of them pretty easily. I know that AT&T was considering doing stock splits to get their price back to about $10 not too long ago specifically to make it attractive to institutional investors.
If you are interested in long term investors only, you avoid stock splits, and keep the price going up. Look at Berkshire Hathaway for an example of this. There shares are worth about $90,000 for the "good ones", and about $6,000 for the "Baby Berks". They specifically never split, and never offer a dividend. It's an interesting model.
If you want to use a single metric to define if a company is worth something, at least use P/E. That's at least something kinda, sorta rational. It takes into account the number of shares, and generally there is an acceptable P/E for any given industry. The P/E of IBM and google could exactly the same, and have IBM's stock at $15, and Google's at $80. You deride that, but any serious investor would realize that the stock price has nothing to do with the value of the company. It's the stock price, and the number of shares that starts to tell you something intelligent. (I believe that number of shares, times the share price is the market capitalization).
I'll point out the truely ironic part, Andrew "Bunnie" Hwang, is the hardcore X-Box hacker who figured out a number of ways to overcome the X-Box security features. A number of slashdot articles have mentioned him, or his projects. Including the facts that he had a really hard time publishing his books on hacks for the X-Box.
Actually, you should read the license carefully some time. There's nothing in one of the licenses about using it to develop a competitor. However, there are two licenses (there might be only one, with two separate conditions, it's been a while since I've read it).
You could just as easily complain that you are required to allow it to "phone home" to the bitkeeper servers and post your source code for the public to see. That's a completely assine requirement if you are paying for it. However, as a non-paying customer who is being granted a "free as in beer" license don't complain too much. He could instead just issue license to Linus and Co, and make the rest of you pay for them, not giving a copy to anyone he doesn't personally know.
You might think he's a real bastard, but you got something from him for free, and now you're complaining that isn't enough? If you had paid for a license, your legal analysis is correct, you do have the right to reverse engineer the product. Larry has flat out said that in publicly posted e-mails.
Don't forget the other important explaination. The plural of anecdote isn't data. It's just like all my friends whom are convinced they are ahead after years of gambling. While it might be slightly true for one of them, even that is highly unlikely unless one of them cheats the house.
People look for those people to have bad years. As a matter of fact, Randy Moss was in there somewhere, and he didn't have such a bad year when he was on the cover (it wasn't as spectacular as his rookie season, but as a receiver it was still a very good statistical year by nearly any measure). Also the SI Cover curse is completely separate from the "Madden Curse".
The SI Cover curse goes back several decades I believe, and is more a matter of being anecdotal. As I recall Lance Armstrong's been on the cover several times during his Tour de France runs, and keeps winning somehow. Normally, it is associated with college football.
People never remember the ones who don't match the pattern. Just like no one remembers all the times they lost $50 at the casino, they can just tell you about the one time left $1,200 up.
Actually, all you have to do is tweak the parameters of the TCP/IP stack. As I recall, Linux as a specific parameter for this. I want to say, it's the transmit window size. They document it as something you only change on a long haul line, which a satallite feed should count as one.
Specifically, you want to allow a lot more packets to be outstanding then a normal TCP connection will allow. This is a bad idea on a low latency connection. It has something to do with windows, and buffering. Also, if you use advanced IP tools to ensure that ACK's get sent before anything else, you'll be much happier.
This thread on the LKML seems to have useful information on it: LKML Thread
Remember the goals here. First, it's they are trying to be binary compatiable with Windows, so applications and drivers will work exactly the same.
In particular the version of Windows they are working on is old, and has been EOL'ed. So you can no longer get support for it. It'd be novel to have a version that was fully compatible, and you could have the source to keep up to date with bug fixes.
Second, they aren't trying to be like Wine. The Wine project is orders of magnitude harder then ReactOS (in some ways). Wine is attempting to make a translation layer from Win32 calls into a Posix/UNIX/Linux environment. That's a whole heck of a lot harder in a number of ways. Things are set absolutely in stone, and can't be changed. On top of all that, at points they get stuck because they are attempting to emulate kernel space functionality in a userspace application a lot of the time.
ReactOS, can make map kernel space things to kernel space things. They can map user space things to user space things. They already have the entire design, and a known model to follow. That's a lot easier then Wine in terms of implementation. Wine is attempting to live withing a much harder set of constraints then ReactOS. However, ReactOS does have to actually implement an entire OS (so it might be a wash). I know I'd rather try and make ReactOS go, then attempt to make Wine work the way it does.
Finally, part of the reason Wine has so many problems, is it started out as a Win3.1 or 3.11 tool, doing 16-bit applications. Now it's moved on to covering a half dozen versions of Windows (at least that many). They also support multiple platforms, and are attempting to be reasonable portable.
The other thing I'll be interested to see is if the ReactOS guys can manage to make it run on MIPS, Sparc, x86, x86-64, Alpha, and any other random platform you can think of. It'd be interesting to see what all they can come up with.
Slashdot Mirror
Some one really ought to tell Microsoft that they can't complete with similar products by charging money. Last I checked, MS Office has probably 2 or 3 legitimate competitors that are completely free. MS Office rakes in cash like very few other products.
While there are plenty of other ways to make money, generally for niche products, charging directly for the software is a good idea.
Kirby
However, in a larger context:
Uhhh, you are taking on a customer for whom you have no tools and no infrastructure for? Who doesn't fit your current model, and fundamentally doesn't fit how you do business? Unless you are laying the ground work to bring in lots more revenue at a lower cost in the future, this might be stupid to do.
Now, a company has to grow, but remember the princepal that says, "Not all customers are profitable". You don't want customers who don't make you money. I remember a story about an advertising company that eliminated 70% of their existing customers and have revenue plumet, but their profits jumped by 30% (as a dollar value, not as a percentage of revenue, they made 2.5Mil instead of 1.5Mil in profit, I believe revenue went from 30Mil to 12Mil).
I know on more then on occasion, the smartest thing the guys in charge where I work is to fire customers. Some customers aren't worth the time or the trouble to deliver service to.
This isn't an anti-Window post, it's merely a matter of considering weather or not this is an area you are planning to expand into, or if this is a one-off, non-scalable solution for a single customer just to get the business.
We run into this quite often, around it's driven by sales people whose sales goals are about bringing in revenue, not bringing in profit. If it costs us $1000 in to bring in $500 in revenue, that's a stupid business proposition. If it's a big chunk of revenue, and you can build it while making money go for it.
Kirby
The third one does, and but only in the context of dragging icons and double clicking them (in the June of 1981 line item). It comes up tangentally later in the 1988 and 1991 sections amount Microsoft. That particular line item in June of '81 I believe is referencing a computer made by Xerox not Apple.
Any chance you'll point out the specifics of the text that clarify that the Mac's didn't specifically take the concept of Icon's from Xerox? Or that Job's inspiration for developing a GUI based computer didn't come directly from his visits to PARC.
Heck, even Jobs openly admits that at PARC, they showed him three things, and he was so blinded by the GUI that he didn't even notice the other two (OO programing, and networking).
http://www.pbs.org/nerds/part3.html
Search for the text "three things". It's right there. Now, Raskin did work there for a little under two years before the PARC visit, but unless it's the "PITS" thing, I don't see anything that leads me to believe that Apple didn't get the idea of a GUI directly from PARC. Raskin might have had the concept in a design 15 years early in his Ph.D, but PARC appears to be the one who shocked Jobs into realizing it was a revolutionary idea. So in the end, it appears PARC deserves a lot of credit you seem to want to deny them.
Bonds give you no authority (in Google's case, that's relatively true anyway). Bonds are also a lot less likely to make any big money for the investors. They'd be a lot better off to privately finance the thing thru a bank. A bond is nothing more then a loan with terms set in a bond where the debt is something you can sell.
There's in theory, low risk, and low reward. It's a different type of investor. An IPO for the company has absolutely no risk. They give up shares (paper) of the company, and they have no liability. Google probably doesn't have the assessts to back up a bond (their value is in their algorithms and the people who run their systems and some data that will be out of date 6 months from the date of purchase, not in physical assests that can be sold during a bankrupty fire sale).
The reason to IPO is to generate income. That's why the company does it. There are other reasons why. It now allows the owners of the company to selectively cash out to a liquid market. So the original investors can get in and out pretty easily. If they didn't want to generate money, I'm going to imagine, they could just get listed as opposed to actually issuing new stock to be sold. Then the original investors could get in and out as they pleased. They could have a much smaller IPO. You do it go generate revenue.
In theory, you do actually have voting rights with Google, it's just that they don't do you any good. At some point, that could actually change, and you still actually get to vote. I'm guessing if every non-founder stock holder votes one way, the founders might consider it.
Microsoft had no dividend for 25 years. I'd be just incredibly happy had I bought them on their IPO. As a long term investment, Google might have similar possiblities. Google I would imagine is going to start accumulating incredibly valuable assests either by creating them, or buying them up.
I don't think that Google will go up in price. They used the same system that the US Gov't uses for selling bonds in order to virtually guarantee a solid price. However, speculative stock buyers along with pre-IPO shareholders who slowly dole out shares while the price runs up. They'd prefer to see Google sell at a $10 price point. There is a lot more run up room. Once the stock moves from 10 to 100, how many people will want in on it? It'll be a feeding frenzy, back like it's 1999.
The person I was responding to, wanted to know why they should care which way it goes (Dutch Auction, or standard IPO). That's why they should care. One way the stock they purchased actually contributed to the value of the company, the other, it contributes to the pockets of a bunch of bankers. If you can get in on it the first day via normal IPO, you'd much prefer that Google use a standard IPO setup (there's plenty of money to be made if you can get in on the low end). If you are a long term investor I know I'd rather have my company get the $3 billion, rather then the investment firm and their friends. I don't own any shares in the investment firm or their friends.
I think Google is silly company to invest in, given that I don't get much in the way of voting rights, and they aren't planning on having a dividend. There's also so serious upside. They could be the next Microsoft. I know I'd be just thrilled if I'd bought every last share of Microsoft I could afford in 1990 and held on for dear life during the ride. It'd be worth about 1000 times what I paid.
Google practically mints cash. They have incredible technology that no one can duplicate. They are insanely popular, and have cornered their market. The problem is that switching search engines is trivial. If someone else can out-google Google, it'll be like when the car was invented. They'll be the best damn buggy whip maker their ever was....
Kirby
When you buy shares of Google, you'd really like Google to get that capital. When you purchase shares of google, you are now an owner in google. It's now in your best interest to be sure that google win's the tug of war between who gets the money. Because it'll maximize google's value.
This isn't so true if you're a speculative buyer who things that Google's price is going to jump up, and if you can just get your hands on it, to turn it over days later while it's on the way up. Then your on the wall street side, and you'd like to see them win.
So yes, depending on the type of investor you are, you have a vested interest in seeing one of the two of them win.
Hopefully, the price won't be the result of playing the games with supply and demand, and the psychology game that happens on Wall Street. There shouldn't be a sky-rocketing value, that if you can get your hands on it, in the first 3 days, and sell hours later a huge profit can be turned.
Liquid markets with stable pricing is good for everyone in the long term. Wall Street's problem is that if your plan isn't going to make money for Wall Street in the short term, they aren't interested. Short sightedness will be the financial ruin of this country if we continue to do things to maximize value in the short run to the detriminte of the value in the long run.
Kirby
In the end, ASCII (which all good e-mail should be presuming it's in English), there is no such thing as a hard linebreak versus a soft linebreak. In desktop publishing such a thing exists. E-mail as a general rule doesn't have a well defined, broadly implemented standard for such a beast.
Which is precisely his beef. It's relatively straightforward a paragraph all on one line into arbitrary length lines (in this case, there are only hard-breaks, and the text viewer implements softbreaks for you). However, it's painful to join things that aren't 80 characters together. The text viewer can't tell if you meant that carriage return to be a softbreak or a hardbreak. On a hardbreak it shouldn't join the two lines, on a softbreak it should. A lot of people aren't terribly consistant (I'm not for example), and their e-mails are a pain to read in an auto-formatted text viewer. I've cringed on more then on occasion when someone quotes my e-mail on a list when they use a 60 char line breaking, and I use 72 char line breaking. Then the last 12-20 characters get split onto their own line. It just looks horrible.
There is some header that can affect this along lines of "Text-flowed" in the Mime Headers, so that things word wrap properly (essentially, your text has soft line breaks, and hard line breaks), where as standard ASCII has no such concept.
I use mutt with vim as my text viewer. I've been known to re-format e-mails on a fairly regular basis and re-save them to be readable to me. I do the interpretation of soft vs. hard breaks and save it so it's easy to read (for me).
Stop being so hard headed and assuming the other guy's just an anal retentive prick. He's got a legitimate point that the interpretation of line breaks is very difficult.
Kirby
In the olden days, people would sell their vote for money. It wasn't until I believe the 1850's or 1860's that we had an anonymous voting system. In an odd coincidence, we imported the Austrialian method back then too!
Before the 1860's you wrote in the name of the candidate you wished to vote for. In small enough precinects, you could literally know everyones handwritting. Before that, you actually walked into the town capital building, and announced your vote in a loud clear voice the the people in charge of keeping track.
Each candidate would have a witness there keeping track of who voted which way, and could then pay off the people who they bought a vote from.
As the other response said, I'd imagine that the first whites to vote for a black in Georgia probably didn't make it too far out of the voting booth before getting harrassed. Unless there was an anonymous system.
Kirby
However, generally when benchmarking one doesn't actually use both the Gbit NIC's (he's a double idiot if he failed to mention this). From the setup, he's got something silly going on. As a general rule, the NIC and the harddrive will need to use up roughly the same about of PCI bandwidth (they are writting roughly the same amount of data, plus or minus framing/headers for the frames/sectors). 140Mbit/s * 2 = 280Mbit per second. So your telling me the "other misc" peripherals are using up 2-3 times the PCI bandwidth of a Gbit card and a harddrive running flat out? What else do you have hooked up to your machine?
More likely, he's running out of CPU, interrupts (they are queuing up faster then they can be serviced), or scalability in the kernel somewhere else, long, long before he's running out of PCI bandwidth. You took the two biggest PCI bandwidth hogs on his machine and showed they used up possibly a fourth, (maybe even a third if you throw in lots of overhead) of the PCI bandwidth, and then claim the "extra stuff", takes up the rest. I'm not buying it.
I'd hear what your saying, if it wasn't just silly. You can't even get close to the Gigabit thru put if you don't have a 800Mhz machine (he doesn't say how fast his machines are). I'm not sure if that uses up all the CPU power of the machine or not.
I've had exactly the problem he describes over NFS with 2.4Ghz Xeon's machines with Gigabit cards on the Northbridge. In the end, it was that NFS protocol has latency in it. In Ethernet parlance, the sliding window is too small. So while I could easily in aggregate get a Gigabit of bandwidth in a number of connections, I could never get a gigabit of bandwidth in a single connection. I wouldn't be shocked if this is also the case with Samba.
I could put about 500Mbit/sec over the cards via NPtcp, but for a standard file transfer over NFS, I couldn't exceed about 14-15Mbyte/sec, which is ~ 120Mbit/sec. You want to get the harddrives, and the PCI contention out of the loop to see what the problem is. Run NPtcp on both ends, check what the maximum thru put is. That'll give you a good number to start with to see how much data your machine can actually push over the wire. From there, it's easier to start creating artificial contention and CPU usage to see what could be causing the problems. If you believe it to be the PCI bus, you can fiddle with the timings of the PCI bus to enhance the bus utilization. Read up on it here: http://www-106.ibm.com/developerworks/linux/librar y/l-hw2.html
It's down at the bottom.
Kirby
I have several harddrives that top out around 14-20Megabytes per second, which turns into roughly the speed rating you are talking about.
I doubt your running out of PCI bandwidth.
It could be the latency, or that you have a poorly tuned network stack. I know that using NFS, getting 12-15Mbit/sec was considered pretty good given the inherient latency of the protocol.
I had similar problems no matter what protocol I was using FTP, HTTP, or scp. What I found was that, I needed to use a network speed tool: NPtcp, which is part of the netpipe tool set.
http://www.scl.ameslab.gov/Projects/old/ClusterCoo kbook/nprun.html
The other thing is to figure out if your cards support Jumbo frames. If they do, it can be a boon to go change your MTU, and modify specific parameters in your TCP stack, and in your application to change the socket options used (specifically, to use a packet size larger then 8k). I'm not sure how to do this under windows, but I've found it readily available under Linux on google searches.
More information is more useful. Knowing what chipset it is based off of, which drivers you are using, what OS, would be mighty helpful to helping you solve your problem.
Kirby
Look how well it worked out for the Autoworkers Unions from the 70's and 80's. They were a powerful group that got absolutely crushed because they thought fighting a free market economy was a good idea.
Now, OpenSource work is good for the economy, if not good for the programmer. It takes highly specialized software and turns it into a commodity. Take a look at it. You can now purchase a computer and load it up with OpenSource software enough to make people highly productive for less then $300 a computer. Take a look 10 years ago. You couldn't hardly get a computer for under $1,000. That's a huge amount of capital that is no longer needed to run a business. Which means the business should have higher profit margins. It lowers the barriers of entry into a marketplace. Lowering the barriers of entry, increases the competition, which forces an increase in the efficiency. That's very good. You want a more efficient economy. It means lower prices for goods, and more capital to start other new ventures with.
Open Source is good for the economy, if not for the programmers. Open Source is good for the economy, in the same sense that roads and bridges are good for the economy. Resources that are generally helpful to everyone, can be built once and only once. That's wat open source is all about. A stable operating system is a necessary tool for almost any office job. A word processer, a web browser, and the myriad of other tools that are broadly used act as a tax on business if you have to pay for them. Which is good for you, if you're the tax collector. It's bad for everyone else. So in the broadest sense, Open Source is bad for programmers, but good for the rest of the world.
Besides which, there will always be a need for customized software which works in specific ways. That process will probably always need highly skilled people who can either write new modules, or extend and configure existing modules. It's sorta like saying, well the car's going to put the horse carriage drivers out of business. Yeah, but they'll be taxi driver jobs coming. It's a cycle, be prepared to adapt and change. Make a lot of money, save a lot of money. Just like, we'd need a lot more programmers to get stuff done if we wrote in assembly. You don't hear anyone clamouring to pass legislation to force people to write in a assembly. Next, you'll be telling me, that John Deer tractors should be outlawed. It's all about efficency. A word processor shouldn't cost money. Society should foot the bill once and make it a public resource. Just like public roads are.
Kirby
Googling around, that's the best link I've seen, but it's relatively short. I haven't looked into in about a year. There is a patch to mount that will allow you to use let you setup cryptoloop (the loopback block device with encryption built in). The maintainer of mount (Andries Bouwer), however wouldn't accept the patch for some legalistic reasons (copyright and/or encryption, I've forgotten the details). Try "man losetup" for information on how to get crypto loop working.
What you do, is have no swap partition, then at the end of the bootup sequence run losetup to setup loopback encryption to either a file or a partition. Then initialize the swap file or partition using mkswap, then swapon /dev/loop0 (or whatever loop device you used).
So something like this:
losetup -e DES /dev/loop0 /tmp/swapfile
mkswap /dev/loop0
swapon /dev/loop0
However, I can't seem to get that working, as my RHEL doesn't have DES as a modules, and losetup doesn't support AES yet. So there's probably a kernel patch to track down. However, it sounds like Mandrake supports this out of the box.
Finally, I'm not much of a 2.6 user, what I described above will work in 2.4. 2.6 I believe either has in the main tree, or it's coming "Real Soon Now", a feature called dm-crypt. dm stands for device mapper. It's the low level replacement for the guts of LVM from the 2.4 kernel. It sounds like a layered/plugin type feature, one of which you can put LVM on top of. Some clever fellows decide to put crypto on top of it. I don't know a thing about that, and have never used it. However, search for dm-crypto in google, here's the most useful page I found in a single query:
http://www.saout.de/misc/dm-crypt/
Good luck.
Kirby
Who exactly is your target audience? People who've never seen UNIX, people who've worked on UNIX environment for years? What is it you are attempting to accomplish with them?
Most of my list would be boring to people who know a lot about UNIX, however some of them are Linux specific.
Personally, I think it's a completely crappy thing. You should get an error telling you: "Attempting to use a constant like a pointer".
So it should print out "10\n10\n";
If that's an ANSI C complier: you should get a warning about no return in main, an illegal declaration of main, and an unknown function printf. You might get away with main, because I believe all functions implicitly return an int.
Kirby
The company owns the copyright on the code your write for them under the legal concept of "works for hire". Unless they failed to pay you for it, they own it. You leaving the building, and letting someone else use it is a copyright violation. The employer you originally wrote it for should be able to successfully sue for copyright violation.
Now, I have plenty of ideas stuck in my head that I've ended up re-implementing in nearly exactly the same way at two different places. I can re-implement my C++ database independent access library in a little under two weeks. I can write the reference counting classes I like to use in a day. I can implement the interface wrapper classes in about 3-4 hours. I can write the socket stream libraries, and the various SSL varients in a day or two. I can write a simple logging scheme in about a day or two. They are written at different times. I can show them as they are constructed, and demonstrate that they are clean room implementations. So I should be in the clear according to what I've read and been advised about the law.
I ended up using the same names, and structuring the code the same way. Some one pointed out I might be breaking the NDA I signed at a former employer. So I ended up re-implementing it all again, this time going out and finding the functionality/API I wanted in publically accessable code (and implementing it from scratch to avoid licensing issues). I documented where each API/functionality idea came from, they we're strewn across several programs. This also showed that the ideas represented "current well known techniques", which no NDA can cover.
You can use exactly the same ideas, and you can implement them at home, and take them into work and let them use them (put in the copyright that you as an individual are the copyright owner, and they are fully licensed to do anything and everything they want with it in royalty free). At least then you'd be legally doing it.
Kirby
Kirby
First off, what difference does it make how many IP's it is? Move 10 DNS names, or moving 100,000 DNS names is pretty straight forward to do via "sed" if you use BIND or most any other text based DNS system. Even if you don't use one, you can change to one by having it be your backup, then use a text editting tool, then switch the roles of primary versus secondary.
Second, the company and the court are forcing this ISP to violate other contracts they have with ARIN.
Third, why doesn't the company have a migration plan? Why can't they take care of it themselves? Migrating services and servers is a pain, but can be done relatively transparently. It's expensive, but I'm not sure how suing your vendor and asking him to pull off nearly impossible technical feats makes this any better.
I have no idea why the ISP has to be complicit in the DNS changes. If they are a hosting company, they should have all of the technical expertise to manage this change without any help from the ISP. All they have to do is continue to pay the original ISP, until they can move to a new ISP.
Finally, any sane person would probably have used mapped IP's so that internally you would have just moved the network and change the mappings at the edges of the network.
The webhosting company is going to become a total screw job. If they had any technical sense, they'd realize that even if the company complies with the requirements and lets the block of addresses be changed, it's still going to be a serious problem with the BGP routing unless they are moving a fairly large block of IP's. If they are moving a large block of IP's, why didn't they just apply for their own? Why don't they apply for their own, and migrate to them in-place? Why is it the ISP's problem to arrange for the technical problems of the client?
Kirby
I believe the author of the article would have two issues with that assertion.
First off, you can have science about how fast grass grows. You have science about how many sexual partners a person has. You have science about how to manipulate people with irrational arguments. Science can be applied to anything that you apply scientific princepals to. Science in a lot of ways, is merely a matter of measuring in a controlled manner and then commenting on such measuring. The usefulness of science is when those measurements are useful and applicable to common every day situations. Like say, your twice as likely to die in a car accident at 50MPH, then 40MPH.
Second, the author sounds like a mathematician, and somewhat of a scientist, and he has a mathematical interest in the filtering of SPAM. It's just as mathematical as using markov chains to model queuing problems to measure how long you'll have to stand in line at the checkout counter. To him, it's an interesting mathematical problem, which in a lot of ways, means that for him personally SPAM classification and the comparison of SPAM classification techniques IS science.
Finally, the results the author is referring to, are due to be published in a peer reviewed journal if I understood it correctly. So in a very technical sense, it is in fact being published for scientific review.
I think a lot of his issue is that you can't use the results of that paper to draw any useful conclusions for yourself if you aren't in a similar situation. As an example, I can get about 18 gallons to the mile in my F150, even though it's only rated for 13/15 city/highway. I manage that by setting the cruise control at the speed right after I switch into the 5th gear, turning off the A/C, driving on predominately flat roads, buying the highest rated fuel, and not stopping for any reason other then purchasing gas. So I could publish a paper saying that a F150 can easily get 18 miles to the gallon. However, that's incredibly useless to anyone who doesn't realize the conditions they have to drive in. His argument is that, the paper doesn't represent the results anyone else would get.
Kirby
It's the ability to flash the BIOS that will make it happen. At some point, Microsoft will have to trust a piece of hardware. If they trust the software, it's merely a matter of time to find out where the branch is that says "yes this is trustworthy", and change the binary so that branch always takes "trustworthy" choice. Just like if I have access to your GPG binary, I can say that a message I sent you is in fact signed by Microsoft (the element of trust everone forgets is that you have to trust the binary sources, in this case, Microsoft can't, as I can fiddle with them). This is an arms race that Microsoft will always lose, it's just a fact of life.
So they must trust a piece of hardware at some point. That hardware must be untamperable, with no way for me to interject myself between it and the Microsoft hardware. As soon as I can interject myself between Microsoft and that piece of hardware, I've won. If I have access to the BIOS, all I have to do is setup some type of virtualization software (Think VMware). At this point, all I have to do is emulate the piece of hardware, and jigger it to always say: "Trustworthy" (essentially a MITM).
If you don't believe that type of attack is plausible, then remember also, I control the client, at some point, I can attack the PKI system. I have access to the PKI portion. At some point, you must have absolute trust of the PKI system, I have the client, what would it take to beat that system? Does Microsoft keep it's list of keys someplace around (it has to, I can subvert that)? That's like giving me access to the root cert's for your Web Browser. You'll trust my hacker sites if I can insert my key into your list of "trustworthy certs". At some point, if I have access to the boot sequence, I can break the system.
The only way it could be secure is to have the hardware have the list of trustworthy keys and have the hardware never give up control to anything that is considered untrusted.
How does Microsoft check that they are running on such a trusted? At some point, they either have to trust the hardware implicitly (which I can fake), or they have it in software that I can modify. At that point, it's either making an untrustworthy piece of hardware (or emulating one), or fiddling the bits of the software. In the end, DRM is a losing proposition. All DRM systems will be broken.
Microsoft might be able to encrypt the software, and only allow it to be decrypted by modules hardware that has the public key embedded inside. However, somebody will just tear the thing apart, or use an X-ray machine to just extract the public key (which at this point is merely a secret piece of data, not really a public key).
Kirby
I bought a truck essentially for the reasons you describe, and now I'm not sure it was the best idea. However, leasing was a very, very good deal for the first 4 years I did. Ford encouraged you to lease heavily. I think the buy vs. lease payments for my first truck was something like: $315 vs $500. Ford priced leasing so attractively, that it was very hard to turn down. They had rigged the system, so that leasing was always a winning proposition (I never paid for full depreciation value of the truck while I had it). Ford was always losing money on it, and hoped to make it up on the used market. The problem, was nobody wanted to pay the inflated used price (to make up for the value I didn't pay for), because it was a cheaper payment to lease then it was to buy.
The buyout price on my first truck was $450/month for 4 years. So I always got more value then I paid for, but theoretically, I wasn't getting the best deal I could have. However, I had a fixed price, full warrantee, other then gas and oil changes, I had no other costs. About the only thing that sucked about it, was you get screwed on miles. You either pay too much or pay too little. On my first lease, I didn't get enough, and ended up owing an extra payment for it. On my second lease, I ended up with an extra 12K miles I paid for, but never used. However, if you are a buy a new vehicle every 6-8 years kind of person, and don't put that many miles on it, buying was just stupid. Over the 6-8 year time frame buying was more expensive then leasing, even over the long term. You have to plan on owning a vehicle for 8-10 years for it to be a money saving proposition.
Leasing was a great deal 4-6 years ago. It was priced too cheaply to not do. At the time, not being able to drive to work was a serious problem. The extra reliability was worth it to me.
Now I'm trying to save money over the long haul, and I live within walking distance of work. So I bought.
Kirby
See, if Google released but a single share, that represented 50% of the company, I bet that single share would be measured in billions of dollars.
If they release 100 Trillion shares, my guess is, fractions of a penny will be the value.
Now it is a good idea to keep your stock price in the $5-$25 price point as it's then a pretty liquid stock, because most investors can afford lots of 100 (generally the smallest unit stocks are sold in by brokers, breaking a lot costs you extra). Institutional investors like pretty liquid stock prices, as they can get in and out of them pretty easily. I know that AT&T was considering doing stock splits to get their price back to about $10 not too long ago specifically to make it attractive to institutional investors.
If you are interested in long term investors only, you avoid stock splits, and keep the price going up. Look at Berkshire Hathaway for an example of this. There shares are worth about $90,000 for the "good ones", and about $6,000 for the "Baby Berks". They specifically never split, and never offer a dividend. It's an interesting model.
If you want to use a single metric to define if a company is worth something, at least use P/E. That's at least something kinda, sorta rational. It takes into account the number of shares, and generally there is an acceptable P/E for any given industry. The P/E of IBM and google could exactly the same, and have IBM's stock at $15, and Google's at $80. You deride that, but any serious investor would realize that the stock price has nothing to do with the value of the company. It's the stock price, and the number of shares that starts to tell you something intelligent. (I believe that number of shares, times the share price is the market capitalization).
Kirby
Interesting. Very intersting.
Kirby
You could just as easily complain that you are required to allow it to "phone home" to the bitkeeper servers and post your source code for the public to see. That's a completely assine requirement if you are paying for it. However, as a non-paying customer who is being granted a "free as in beer" license don't complain too much. He could instead just issue license to Linus and Co, and make the rest of you pay for them, not giving a copy to anyone he doesn't personally know.
You might think he's a real bastard, but you got something from him for free, and now you're complaining that isn't enough? If you had paid for a license, your legal analysis is correct, you do have the right to reverse engineer the product. Larry has flat out said that in publicly posted e-mails.
Kirby
People look for those people to have bad years. As a matter of fact, Randy Moss was in there somewhere, and he didn't have such a bad year when he was on the cover (it wasn't as spectacular as his rookie season, but as a receiver it was still a very good statistical year by nearly any measure). Also the SI Cover curse is completely separate from the "Madden Curse".
The SI Cover curse goes back several decades I believe, and is more a matter of being anecdotal. As I recall Lance Armstrong's been on the cover several times during his Tour de France runs, and keeps winning somehow. Normally, it is associated with college football.
People never remember the ones who don't match the pattern. Just like no one remembers all the times they lost $50 at the casino, they can just tell you about the one time left $1,200 up.
Kirby
Specifically, you want to allow a lot more packets to be outstanding then a normal TCP connection will allow. This is a bad idea on a low latency connection. It has something to do with windows, and buffering. Also, if you use advanced IP tools to ensure that ACK's get sent before anything else, you'll be much happier.
This thread on the LKML seems to have useful information on it: LKML Thread
Kirby
In particular the version of Windows they are working on is old, and has been EOL'ed. So you can no longer get support for it. It'd be novel to have a version that was fully compatible, and you could have the source to keep up to date with bug fixes.
Second, they aren't trying to be like Wine. The Wine project is orders of magnitude harder then ReactOS (in some ways). Wine is attempting to make a translation layer from Win32 calls into a Posix/UNIX/Linux environment. That's a whole heck of a lot harder in a number of ways. Things are set absolutely in stone, and can't be changed. On top of all that, at points they get stuck because they are attempting to emulate kernel space functionality in a userspace application a lot of the time.
ReactOS, can make map kernel space things to kernel space things. They can map user space things to user space things. They already have the entire design, and a known model to follow. That's a lot easier then Wine in terms of implementation. Wine is attempting to live withing a much harder set of constraints then ReactOS. However, ReactOS does have to actually implement an entire OS (so it might be a wash). I know I'd rather try and make ReactOS go, then attempt to make Wine work the way it does.
Finally, part of the reason Wine has so many problems, is it started out as a Win3.1 or 3.11 tool, doing 16-bit applications. Now it's moved on to covering a half dozen versions of Windows (at least that many). They also support multiple platforms, and are attempting to be reasonable portable.
The other thing I'll be interested to see is if the ReactOS guys can manage to make it run on MIPS, Sparc, x86, x86-64, Alpha, and any other random platform you can think of. It'd be interesting to see what all they can come up with.
Kirby