Google will start adding similar revenue-generating ideas, or their financial
backers will start demanding changes.
Two-thirds
of Google's revenue is from ads. They are opening new
sales offices (e.g. Germany), but slowing down tech hiring. That suggests they are betting on increasing
ad revenue at a time when their competitors have decided that ads alone can't sustain
search-engines. Google's techie hiring cutback also suggests that
they don't think additional software R&D can help them grow as much
as investing in non-tech areas. [Estimates I've seen of Google's revenues are US$30M - $70M a year,
with their CEO saying that makes them just about profitable.]
Worse for Google, they hold few patents for their basic technological advantage, and their infrastructure (including their huge database) could be rebuilt in a few weeks by a cash-rich M$. The only protection they have against
Teoma et al is their staff -- but loyalty can be bought. (Google uses options to encourage employees to stay. If the options cease to look promising, some people will leave.)
Another problem facing Google is their staff itself. 50 of their 250 employees are
PhD's. That means they have lots of valuable
technical knowledge, but it also means that 50 of their highest-paid
employees have a collective 0 years experience in business planning. Consider that their
senior management lacks a CFO at all, and is loaded with CS doctors
who tend (like normal geeks) to want to work on "cool" things instead of profitable ones.
Google's proud of its lack of advertising -- but don't they also lack the marketing that would
produce such advertising? Look at two of recent new products: the USENET database
(cool, but what good does it do for *Google*?), and the shopping-catalog database
(a possible money source...but very risky, requiring licensees to share their
revenue stream and catalog-shoppers to change
their habits.)
Being private means Google can avoid stockholder demand for quick profits...sort of.
Their only source of funds is two VC firms, since the founders had little money of their own.
The two firms [1][2]-- each of whom has a seat on Google's board -- will eventually demand return on their
$25 million investment. Remember, the folks who gave Google its money want to see
profits, and have *lots* of experience in tweaking start-ups to generate them.
Don't get me wrong -- Google's great;Brin & Page deserve copious kudos & cash.
However, I'm watching for some danger signs:
Lots of new "Sales" or commission-based positions at company
An exodus of employees. (With their high retention rate, "exodus" might mean 10 people.)
Research efforts into non-Linux infrastructure.
A lot of new product offerings that target consumers directly.
I'm also watching for signs I'd consider *good*:
A removal of one (or both) founders from day-to-day operations.
More parterships with content producers.
Another level of financing (demonstrating VC belief that they can grow.)
The plan to base product engineers' raises and bonuses on their code's quality will encourage programmers to write better code...but it's not enough to lead to safer Microsoft products. The problem is that manager / executive bonuses at M$ are still based on product profits, and are generally given as stock options.
This means the managers will still target profitablity over security.
..."unbreakable" doesn't really mean unbreakable, or something...
Oracle said that 9i "is unbreakable". As President Clinton could easily tell you, the key word here is 'is'.
If companies faced lawsuits and financial penalties when vulnerabilities were found and exploited, they would strongly discourage white-hat hacking, independant vulnerability testing, etc. It would be in Microsoft's best interests to immediately sue anyone who reports a flaw. (White hat hacking violates US law just as black hat does.)
Lawyers would start to be accused of Bugtraq chasing.
Hamidi was cited for "trespass to chattel", which means he (according to Intel) used Intel physical property [mail servers] for his own gain without permission. It's a legal term only slightly related to trespassing.
It's equivalent to "borrowing" your neighbor's car while they're on vacation, but refilling the gas tank when you put it back. Nothing's really lost, but you got value out of someone else's stuff.
In Intel v. Hamidi, the appeals court said that the volume of email was important in deciding whether there's an offense, and specifically said that sending 1 email to an email address is not trespass to chattel. They didn't say where the line was.
The decision
also available here from the court itself said the lower court erred in not letting Ferguson pursue tresspassing charges against the spammer, despite his evidence that physical damage was done to his computer. I admire his creativity: the physical damage he cites is the fragmentation of his hard drive due to the spam in his inbox.
Has he also considered the physical damage that can occur by increasing your ejaculate by 581% and shooting it 13 feet?
teridon's danger seems even worse that the Brit problem, because he's dealing with science satellites, which release more info to the public than do military ones. His user info suggessts he's discussing the
SOHO
(Solar & Heliospheric Observatory) satellite, which has already demonstrated a hacker-desirable feature:
a buffer overflow in code that caused control problems in the satellite.
Uplink/Downlink details on SOHO are readily available, e.g.:
Uplink Frequency(s) 2067.271 MHz
Downlink Frequency(s) 2245.000 MHz
Commands: 16Khz subchannel @ 2kbps
Uplink transmitters used: Gladstone, Canberra, Madrid
..IMHO it would be feasible to decipher the command structure, especially because descriptions of the commands being used are published on NASA "project home pages". Encryption would help, but would have to be extended to cover the the networks that author the commands, i.e. Goddard, JPL, etc.
Good article, especially the comments about the cost of support staff for Linux. I doubt the "core problem with Linux is that you've got to work hard to connect USB drivers", though.
I've sold both Windows and Linux based tools for awhile, after after a lot of soul-searching (and checkbook-balancing), I've decided I prefer selling Windows solutions. Four reasons --
It's easier to offload Windows work to others because there are more people capable of doing basic tasks on Windows than can do it in Linux.
Windows offers better profit margins...It's easier for me to mark up $1000 software by 10% than it is for me to charge a $100 "price" for free software.
There's more training available for Windows tools than Linux stuff. Linux is mostly learn-as-you-go or learn-from-a-friend. I hate it, but local training companies are bursting with classes, books, and testing for MS stuff.
It's what my customers ask for. People start conversations with me all the time by asking, "Can you handle this Windows issue..." I almost never hear "Can you help me with this Linux problem..." (possibly because, as the article points out, Linux users aren't as technically challenged as MS users.)
I still prefer Linux for *my* stuff -- I just like Windows for *other people's* stuff.
At least, according to
this timeline, we
might need thank Tolkien for more than
lembas. Intrigued by
Google's
USENET archive, I tried to hunt down the origin of the word "spam".
EFF
and Wired both give the
party-line answer: the word derives from MUDs (Multi User Dungeons) of the late 80s to describe
"unwanted stuff", and came from the Monty Python spam sketch.
The USENET posts I found, though, flesh out the story a little.
The origin seems tied specifically to TinyMUD, written by
Jim Aspnes, inspired partly by
Zork and
earlier PDP-10/11 MUDs. TinyMUD was launched in August of 1989.
TinyMUD's advantage over other MUDs was that visitors could not only wander around
a dungeon (think "maze of twisty passages, all alike"), but they could also add new rooms and
monsters on the fly.
Searching USENET, it seems there were two meanings of the term "spam". One definition
was based on people abusing the ability to add new objects to the TinyMUD world:
April 17, 1990, posted
by Jon Blow: ...By this time, the wizards [dungeonmasters] had locked off a few areas that were just
spam-for-the-senses...
June 27, 1990, posted by Vintage Mutant Ganja Technerd: For example, a delay of 5 to 10 seconds between object creations and logging in, will all do the trick of 'limiting' spamming without the juggling of quotas, login times, keeping track of hosts, et al.
October 4, 1990, posted by A Molitor: ...when you run a MUD advertised as
having few or no rules, a MUD where you can do anything, players *will* spam
it. This is not conjecture, but documented historical fact. Ask around about
BloodMUD some time.
However, the second meaning of the word, and the one that seemed to appear earlier in
USENET, is the one that more closely resembles the meaning we use today:
From: Jon Blow (blojo@tornado.Berkeley.EDU) Subject: Re: Word wrap
Newsgroups: alt.mud
Date: 1990-01-22 23:18:55 PST
Right now my entire adventure is formatted to be easy-readable in 80
columns. This is also a pain, since 1) It takes much longer for me to write it,
and I constantly feel a loss of artistic quality when I am forced to reword so
that a line will fit; 2) People with wordwrap must turn it OFF, or the adventure
will look like Spam. Bummer.
Other posts (and various MUD histories on the net) discuss the problem of MUD
visitors who used various commands (most often the 'say' command) to fill other people's
screens with unwanted text, thus scrolling more important things off the screen.
The first place I found the word "spam" being applied to USENET posts themselves was
here,
related to a bot that accidentally regurgitated other posts in the news.admin.policy newsgroup.
Since most MUD Histories
attribute their rise to the fantasy genre of Tolkien (and to a lesser extent Dungeons
and Dragons), don't forget to thank Middle-Earth (and 25-line CRTs) for 'spam' when you
see the movie
next week. There are doubtless other etymologies; I'm just basing this on the only evidence I found.
As a side note, to Google employees the term "spam" refers not to unwanted email but rather to the underhanded tricks folks try to boost their search-engine rankings.
Hardware is fine and dandy, but have you ever tried to find the right drivers for a Turing Machine? ArcServe and Veritas drivers only support a few hundred gigabytes of tape.
You'd think after 60 years someone could have at least invented a Turing Autoloader.
Make me something that doesn't suck,and I'll pay for it, don't force me to upgrade every 20 minutes to a more bloated piece of crap...
Unfortunately, if I write software that doesn't suck, doesn't need patches, and does what you want, you'll buy one copy (Netware 3, WinZip, Eudora) and in 2 years I'll be bankrupt.
If I write software with tons of broken features and requiring constant upgrades for 'compatibility' and security (SAP, QuickBooks, and Windows 95), I'm guaranteed plenty of repeat customers.
Now if you'll excuse me, I need to go buy a $100 ink cartridge for my $30 printer.
Slashdot Mirror
What's the best way for genuine, qualified, informed candidates to distinguish themselves from this rabble?
Hack Slashdot, then brag about it.
Two-thirds of Google's revenue is from ads. They are opening new sales offices (e.g. Germany), but slowing down tech hiring. That suggests they are betting on increasing ad revenue at a time when their competitors have decided that ads alone can't sustain search-engines. Google's techie hiring cutback also suggests that they don't think additional software R&D can help them grow as much as investing in non-tech areas. [Estimates I've seen of Google's revenues are US$30M - $70M a year, with their CEO saying that makes them just about profitable.]
Worse for Google, they hold few patents for their basic technological advantage, and their infrastructure (including their huge database) could be rebuilt in a few weeks by a cash-rich M$. The only protection they have against Teoma et al is their staff -- but loyalty can be bought. (Google uses options to encourage employees to stay. If the options cease to look promising, some people will leave.)
Another problem facing Google is their staff itself. 50 of their 250 employees are PhD's. That means they have lots of valuable technical knowledge, but it also means that 50 of their highest-paid employees have a collective 0 years experience in business planning. Consider that their senior management lacks a CFO at all, and is loaded with CS doctors who tend (like normal geeks) to want to work on "cool" things instead of profitable ones.
Google's proud of its lack of advertising -- but don't they also lack the marketing that would produce such advertising? Look at two of recent new products: the USENET database (cool, but what good does it do for *Google*?), and the shopping-catalog database (a possible money source...but very risky, requiring licensees to share their revenue stream and catalog-shoppers to change their habits.)
Being private means Google can avoid stockholder demand for quick profits...sort of. Their only source of funds is two VC firms, since the founders had little money of their own. The two firms [1][2]-- each of whom has a seat on Google's board -- will eventually demand return on their $25 million investment. Remember, the folks who gave Google its money want to see profits, and have *lots* of experience in tweaking start-ups to generate them.
Don't get me wrong -- Google's great;Brin & Page deserve copious kudos & cash. However, I'm watching for some danger signs:
- Lots of new "Sales" or commission-based positions at company
- An exodus of employees. (With their high retention rate, "exodus" might mean 10 people.)
- Research efforts into non-Linux infrastructure.
- A lot of new product offerings that target consumers directly.
I'm also watching for signs I'd consider *good*:- A removal of one (or both) founders from day-to-day operations.
- More parterships with content producers.
- Another level of financing (demonstrating VC belief that they can grow.)
Whew! (my $rant->time_complete=now();)The plan to base product engineers' raises and bonuses on their code's quality will encourage programmers to write better code...but it's not enough to lead to safer Microsoft products. The problem is that manager / executive bonuses at M$ are still based on product profits, and are generally given as stock options.
This means the managers will still target profitablity over security.
..."unbreakable" doesn't really mean unbreakable, or something...
Oracle said that 9i "is unbreakable". As President Clinton could easily tell you, the key word here is 'is'.
If companies faced lawsuits and financial penalties when vulnerabilities were found and exploited, they would strongly discourage white-hat hacking, independant vulnerability testing, etc. It would be in Microsoft's best interests to immediately sue anyone who reports a flaw. (White hat hacking violates US law just as black hat does.)
Lawyers would start to be accused of Bugtraq chasing.
If everyone in Belgium had a copy of XP, police could just require you to carry your Microsoft Product Activation Key with you at all times.
...that the FBI would announce Magic Lantern, and two weeks later SETI releases an "improved" version of its message.
Hamidi was cited for "trespass to chattel", which means he (according to Intel) used Intel physical property [mail servers] for his own gain without permission. It's a legal term only slightly related to trespassing.
It's equivalent to "borrowing" your neighbor's car while they're on vacation, but refilling the gas tank when you put it back. Nothing's really lost, but you got value out of someone else's stuff.
In Intel v. Hamidi, the appeals court said that the volume of email was important in deciding whether there's an offense, and specifically said that sending 1 email to an email address is not trespass to chattel. They didn't say where the line was.
The decision also available here from the court itself said the lower court erred in not letting Ferguson pursue tresspassing charges against the spammer, despite his evidence that physical damage was done to his computer. I admire his creativity: the physical damage he cites is the fragmentation of his hard drive due to the spam in his inbox.
Has he also considered the physical damage that can occur by increasing your ejaculate by 581% and shooting it 13 feet?
Uplink/Downlink details on SOHO are readily available, e.g.:
At least this is evidence that MS can't sneak in the a backdoor on XP...Otherwise they'd just patch everyone and skip the media fallout.
n /MS01-059.asp
Here's the MS bulletin with links to the download: http://www.microsoft.com/technet/security/bulleti
FYI, for the latest patches, don't use windowsupdate.microsoft.com's automatic-upgrade detector...It takes a couple of weeks before patches make it into that system.
I've sold both Windows and Linux based tools for awhile, after after a lot of soul-searching (and checkbook-balancing), I've decided I prefer selling Windows solutions. Four reasons --
I still prefer Linux for *my* stuff -- I just like Windows for *other people's* stuff.
The story must be true, because I found a bug in XP.
.Net projects, too.
Based on my observations, Mr. Afroze seems to have worked on the Win 95/98/2000, Outlook, SQL Server, Office, VB Studio, and
EFF and Wired both give the party-line answer: the word derives from MUDs (Multi User Dungeons) of the late 80s to describe "unwanted stuff", and came from the Monty Python spam sketch.
The USENET posts I found, though, flesh out the story a little. The origin seems tied specifically to TinyMUD, written by Jim Aspnes, inspired partly by Zork and earlier PDP-10/11 MUDs. TinyMUD was launched in August of 1989. TinyMUD's advantage over other MUDs was that visitors could not only wander around a dungeon (think "maze of twisty passages, all alike"), but they could also add new rooms and monsters on the fly.
Searching USENET, it seems there were two meanings of the term "spam". One definition was based on people abusing the ability to add new objects to the TinyMUD world:
For example, a delay of 5 to 10 seconds between object creations and logging in, will all do the trick of 'limiting' spamming without the juggling of quotas, login times, keeping track of hosts, et al.
However, the second meaning of the word, and the one that seemed to appear earlier in USENET, is the one that more closely resembles the meaning we use today:
Subject: Re: Word wrap
Newsgroups: alt.mud
Date: 1990-01-22 23:18:55 PST
Right now my entire adventure is formatted to be easy-readable in 80 columns. This is also a pain, since 1) It takes much longer for me to write it, and I constantly feel a loss of artistic quality when I am forced to reword so that a line will fit; 2) People with wordwrap must turn it OFF, or the adventure will look like Spam. Bummer.
Other posts (and various MUD histories on the net) discuss the problem of MUD visitors who used various commands (most often the 'say' command) to fill other people's screens with unwanted text, thus scrolling more important things off the screen. The first place I found the word "spam" being applied to USENET posts themselves was here, related to a bot that accidentally regurgitated other posts in the news.admin.policy newsgroup.
Since most MUD Histories attribute their rise to the fantasy genre of Tolkien (and to a lesser extent Dungeons and Dragons), don't forget to thank Middle-Earth (and 25-line CRTs) for 'spam' when you see the movie next week. There are doubtless other etymologies; I'm just basing this on the only evidence I found.
As a side note, to Google employees the term "spam" refers not to unwanted email but rather to the underhanded tricks folks try to boost their search-engine rankings.
Hardware is fine and dandy, but have you ever tried to find the right drivers for a Turing Machine? ArcServe and Veritas drivers only support a few hundred gigabytes of tape.
You'd think after 60 years someone could have at least invented a Turing Autoloader.
Make me something that doesn't suck,and I'll pay for it, don't force me to upgrade every 20 minutes to a more bloated piece of crap...
Unfortunately, if I write software that doesn't suck, doesn't need patches, and does what you want, you'll buy one copy (Netware 3, WinZip, Eudora) and in 2 years I'll be bankrupt.
If I write software with tons of broken features and requiring constant upgrades for 'compatibility' and security (SAP, QuickBooks, and Windows 95), I'm guaranteed plenty of repeat customers.
Now if you'll excuse me, I need to go buy a $100 ink cartridge for my $30 printer.