> Or as a compromise, set the firewalls (home and > work) to only accept connections to port 22 from > specified IP addresses
I suppose I could get into my home system from work that way, but my home system is on dynamic IP, so I still wouldn't be able to get into work from home, if I go that route.
It does make sense. The math behind why has been pointed out by others, so I'll just offer an analogy...
Is it faster to drive your car, or walk?
Depends how far you're going. If you're going several miles, it's going to be a lot faster to drive. If you're going across the street, it's faster to walk, because you avoid the overhead of going out of your way to where the car is parked, cleaning the snow off the windshield, unlocking and starting the car, parking the car, and getting out. When I was in high school, I walked. Once a friend who was just arriving offered me a ride from the end of my driveway. We ended up parking further from the school door than where he picked me up.
The O(1) scheduler is the car. It'll get you to that college in the next state faster, but it's no help for getting across the street.
> The only version available is for Internet Explorer for Windows.
googlebar.mozdev.org
Unfortunately, this doesn't seem to install properly on current versions of Mozilla, but it worked great with 0.9.9, and I suppose the bitrot will be fixed eventually.
I would _use_ that. Currently to transfer things from one OS to the other on a WinXP/Linux dual-boot system is a pain (I'm using an smb share on a different PC on the network...)
Yeah, I know WinXP can theoretically use FAT, but I don't (call me crazy) particularly want to have to reinstall it, and it came preinstalled on NTFS.
> His new role plays to perhaps his greatest skill--that > uncanny ability to foresee how emerging software > technologies can be woven together and parlayed into > must-have "industry standard" products, which, in turn, > reinforce demand for other software from Microsoft
> If Longhorn really does turn out to be a Super Windows--a > big if--it will handle so many functions of computing that > Oracle, Sun, AOL Time Warner, and Sony may find themselves > with less to do...
> Because Gates' geeks are completely overhauling the > operating system, they'll also have to redesign most of > the company's other software products and services to take > full advantage, including the MSN online service, its > server applications,...
If I read that right, we're saying three things:
1. Bill Gates' job is to interweave (i.e., make
interdependent) previously distinct technologies
so that Microsoft products are must-have.
2. Oracle, Sun, AOL, and Sony are targeted for
takeover. We already know the database filesystem
will make life tough for Oracle. So, what is the
threat to Sun, AOL, or Sony? I think I can answer
about AOL (below).
3. Read that third quote again. And again. Does
that sound to you like the MSN internet service
will have to be changed in order to be compatible
with the new MS OS? Does that mean, then, that
_only_ (the revised) MSN will work as an internet
service for users of the new MS OS?
Remember what was said in the Holloween Documents about decommoditizing protocols? This is it; either I've misread something badly, or Bill Gates wants to "decommoditize" internet service. We already know MSN users have to use Outlook; in a 2006, users of this OS will have to use MSN (and, of course, Outlook).
Fortunately, 2006 may be too late. In 1998 Unix was not ready for the desktop. In 2002, Unix *is* ready for the desktop. By 2006, I expect Microsoft to have lost some of that market share to Linux and OS X. With any luck, Longhorn will be too late to lock down the whole market for them.
That's how I read it too. He's saying that if you install the update and use privilege separation, you won't be vulnerable during the time between when the bug and true fix are published and the time you get them installed.
I'm thinking about going another route, and turning setting things up so sshd only listens for connections from the local network until after the bug and fix are published and I get my systems patched. That loses me the ability to login to my home system from work, or work systems from home, during the interim, but it lets me wait for a stable fix.
Wireless networking is going to be a while yet competing with wired networking. Except maybe for line-of-sight wireless (which isn't really the same thing; it may not have copper wires, but it has lines the information has to follow sure enough), wireless just isn't developed to the point where it can offer as much bandwidth as similarly priced non-wireless options. The satellite stuff that's supposed to compete with DSL for consumers costs twice as much per month and has preposterous installation fees. AirPort and similar local wireless options are dog slow compared to even vanilla 10BaseT ethernet but cost more like gigabit ethernet.
I don't think tech-savvy persons are more *or* less likely to pirate music. I think the two traits are orthogonal. Personally, I don't pirate music, but I also don't buy this guy's stuff, because it's entirely out of genre for me.
I almost never listen to 20th century music, except for the occasional Yankovic or Shickele, or a little polyphonic (non-monodic) a capella stuff, or stuff other people select and play when I happen to be present. Mostly given the choice I listen to baroque (especially late baroque), and sometimes a little romantic (in the traditional sense; Chopin or whatnot), or _occasionally_ some of the better classical (I'm not into Mozart; Dvorak is pretty cool though). I tire of monody (one lead melody part with support from parts written around it) quickly and have a marked preference for real polyphony (interwoven separate but equal parts designed to go together) or fugue. So as you can imagine I have no motivation to pirate anything produced by Perl Jam.
Now, I'm not suggesting that tech savvy people are unlikely to listen to modern music. What I am going to suggest is that tech-savvy users have very specific ideas about what music they like and will pay for, and are less likely to buy an album just because it is produced by a popular group, even a group that has formerly produced albums they like. Part of what makes a geek geeky is that he gets adamant about small things other people don't seem to care about. A geek will refuse to pay for something he does not want, on principle, even if it's considered fashionable outside of geek circles. (Unless it's a technical gizmo he can mess with and reprogram, in which case some geeks will crawl naked over a field of glass caltrops to buy it, but nevermind; music does not fit that category.) But I don't think geeks exhibit a marked tendency to pirate, or not to pirate, any more than the rest of the population at large.
Now, people who listen to baroque are probably less likely to pirate music illegally than people who listen to modern music, but that's a different matter. (Think in terms of, lesse, 2002 less seventy is 1932... The artist would have to be, err, 180 years old or so. Yes, the performances are copyrighted, but the lack of composer royalties drives the prices down a LOT. Plus, the ecconomy of scale is quite significant for some of it. Bach for example probably sells more albums every year than this Moby guy has sold in his life (though perhaps not more _dollars_ worth of albums).)
I also have had good experiences purchasing from tcwo.com
They ship promptly and were good about taking a personal check (since I don't do credit cards) given adequate ID (a driver's license number). All-around good experiences. I've purchased from them several times.
Their selection is not exhaustive, but it is a good representation of the things you are likely to need. Most of the things I've needed that they didn't have were specialty items that I ended up getting direct from the relevant manufacturer (an Avant Stellar keyboard from CVT, for example), although once when I needed a lengthier cat5 patch cable I had to go elsewhere (Computers4Sure IIRC) for it.
I had a different experience (one I would tend not to charactarise as positive) attempting to purchase from LinuxWarehouse, wherein I was sold an item that was no longer in stock, and my order was cancelled without notification, but my check was cashed... I eventually got my refund, but it took months. (Lest this be mistaken for libel, I can provide specific details by email upon request, but I won't bore all of slashdot with them.)
I think the etymology of the term will best answer your question.
The original thinking was that there are not separate plural and singular forms. This goes along with (part of) why the term "spam" was chosen to represent the stuff. Like the meat, there are no clear divisions between pieces; it's all just _stuff_.
Think about it. If your mailbox contains 100 messages (yeah, you wish it were so few...), and all of them are spam, and 100 other people (again, never so few...) like you also received 100 messages, all spam, and without sophisticated IP subnet tracking none of you can easily tell who sent any given message, and you're all receiving messages from a different subset of the set of all spammers, and each spammer sends you a different subset of his set of all messages...
So, any given message you received is very similar to a random subset of the other ones you receved _and_ was also received by a random subset of your 100 friends, each of whom also has a random collection of messages that are very similar to this one, but may not be exactly the same as the ones you have...
What you have (all of you, collectively) is a messy conglomeration of mixed parts. Like the Hormel product.
There is no important semantic distinction between one message and the next, or between the same message received by different people, or sent using different relays... it's not generally useful to talk about an individual message. Collectively, it's all just spam.
When you want to start filtering it, you (or your filtering agent) have to look at each message individually. Especially if you're building a database of known spam. But otherwise, it's all just spam.
There were at one time some other terms to go along with spam: jello and velveeta come to mind. The thinking was that junk posts to usenet were different from junk email, and that multiposts were different from crossposts, and multiposts to one newsgroup repeatedly different from (nearly) simultaneous multiposts to multiple groups, and so on and so forth. These distinctions turned out to be unimportant, largely because the same stuff gets sent in all of the above ways, and so now it's all just spam.
This is a good thing. Most people have very vague notions about
what Windows and Microsoft even are; the advertising campaigns
have some of them believing that it's an important part of
computing, but only because all computers seem to have these
buzzwords "Windows" and "Microsoft" attached to them. Exposure
to the idea that computers don't need these buzzwords is a
good thing.
In terms of user experience, users want to do a few simple
things:
Send and receive email, without understanding anything
about how email works. When someone sends them an
image attachment, they want it to just display for
them, and have a clearly visible "print" button.
They also want to be able to exchange inane animated
greeting cards (the ones spammers deploy to collect
email addresses), so they'll need the Flash plugin.
But they don't have even the foggiest notion what
a "plugin" is, and they shouldn't have to.
Print stuff out. This means mostly pictures,
bog-standard word-processing documents (letters,
resumés, garage sale signs,... nothing
complicated), and the kind of thing people used
to use Print Shop for in the 80s (mostly inane
greeting cards with cheesy clip art, but these
days they want to do this in color; banners are
also popular). Printing pictures is no problem.
Word Processing is no problem; Open Office is
serious overkill for these people. The thing
that remains in this category is the cheesy
greeting-card/certificate/banner printing
package, and I've discovered that people will
crawl over broken glass to do this stuff. The
software can be _horrific_ (a la Print Artist)
and they'll _LOVE_ it. Quality is not necessary,
and ease of use is really not important either,
as long as it will let them insert stupid
clip art and style bits of text with shadow
and outline effects and stuff, and give them
prefab templates to modify. Currently I don't
know of a Linux app that fits this bill, but
maybe that's because I wasn't looking.
Surf the web. This shouldn't be a problem. I've
been deploying Mozilla for a while now at a public
library, where the people who use it have no PC
at home and know virtually nothing, and Mozilla
works fine; I get very few complaints, and those
I do get have to do with printing or with the
difficulty of navigating certain sites.
Play silly little games. Not a problem. Give
em a dozen kinds of Solitaire, Gnome Mines, Iagno,
and a handful of others, and they'll be happy
playing them quite literally forever. (Yes,
there are also people who want cool games, new
games, 3D shooters, and such, but those people
are younger and know more about computers.)
That's pretty much it. Most people don't know
they can do more than this with computers.
I'm glad Wall-Mart is no longer claiming that
LindowsOS runs most MS programs. Lindows was
not ready for that claim. But Linux *is*
ready, or very close to ready, for the consumer
desktop, as long as it comes preinstalled and
preconfigured. I worry just a little about
the silly-greeting-card thing... developers
don't do such inane things, and I don't know
whether anyone has put together a Print Artist
equivalent for Linux.
First, I want to say that I don't consider dead tree copies to be useless in all cases. There are some books I'd actually rather have in dead tree form -- mostly ones that I'm never going to need to grep but will want to take to the bathroom. For example, _Just for Fun_ (by Torvalds), or Knuth's book on surreal numbers. Other books (any reference book most programming books) I'd rather have in electronic form.
But I really want to address the eye strain point.
Eye strain can be solved by changing your colour settings and otherwise tweaking your environment.
Black on blinding white will of course cause eyestrain; even most dead tree books that are meant to be read for hours on end aren't printed on white paper, but a softer off-white shade. More significantly, paper is an absorption/reflection medium; it doesn't emit light. On a computer screen, you want a darker background and a lighter foreground, if you want to sit in front of it for days on end without eyestrain. Also, a common mistake is to use the highest available contrast. That's great for road signs, where you want to get the reader's attention immediately, but for long exposure a somewhat lower contrast is better. (Not _LOW_ contrast, of course -- that's even worse than really high contrast.) The best combination I've found so far is wheat (about #F7DFB5; or so) on slate green (#305050;). Soft tertiary colours are easy on the eyes. Yes, easier than paper, if you have a good refresh rate on your monitor. Having incandescent lighting (instead of fluorescent) helps too.
There's a Gnome/GTK theme, called thEmacs, that works pretty well for me on my current (Mandrake distro, Gnome session) system. For Windoze, you can do something similar pretty easily, and I have. (For W95, you can get my wheat on slate green and a couple of other colour schemes from my website (click on "Personal", then on "Colour Schemes I have Perpetrated" -- the base URL is in my/. profile); that may not work as intended on other versions of Windows, since gradients and things were introduced in Windows 98 the format for colour schemes in the registry may have changed).
Oh, and for OO design and reading enjoyment...
on
General IT Books?
·
· Score: 1
The Inform Designer's Manual, by Graham Nelson. 4th edition if possible.
Dead tree copies of this one are slightly difficult to find in stores, but this book is worth going out of your way to get. It's not just loaded with interesting examples of OO design in action -- it's also one of the most _interesting_ programming books ever written. The Inform language is niche, but this book is worth reading for its own sake.
I'm sure it was so obvious you all just forgot to mention it, that you all have it on your shelves both at work and at home, but it really needs to be said anyway, in case there's some poor newbie who just doesn't know...
You've gotta have the Hitchhiker's Guide to the Galaxy.
I'd also like to add that the Camel Book (Programming Perl, already mentioned) is one of the best of the lot.
The browser war has only just barely begun. When MS first announced IE, nobody took them seriously. Netscape bumbled the first battle, and fell back to regroup. But the war is about to begin in earnest soon. You can see the signs, if you pay attention... AOL is getting ready to fight with all the (considerable) weapons at their disposal. Including...
* Bundling. They're testing both IE and Gecko technology
for their next release. I posit that they're deciding
whether it's time, and what the response will be if they
switch. They've lost their AOL-on-desktop deal with
MS, so they _want_ to switch, but what will the user
response be? They're waiting to see.
Why do I think they're trying to decide whether it's
time? Because you can watch them guaging user response,
as they do tests with the new technology, distribute
it among a smaller user base (Compu$erve), and so on.
They're watching to see how it is received. Netscape
6 was received poorly, and they waited. Netscape 7,
now in beta, already has better response than 6. Don't
think they won't notice.
When will it be time? IMO, soon. Mozilla's user
experience shaped up considerably over the winter;
something happened: people who tried 0.9.7 and
0.9.8 liked it enough not just to use it themselves,
but to recommend it. When 0.9.9 came out, mozilla.org
had to get mirrors and more bandwidth, to handle the
increased demand. Sure, these are all the lunatic
fringe, people who will try a new technology before
it's really popular. But here's my point: Netscape
6 was branched before all of this, and Netscape 7
comes after. There has been a fundamental shift,
meanwhile, in how it is received. AOL is watching
the response, and they're going to see that the
response that comes back from Netscape 7 will be
good. It will be time. That's my prediction.
* Media coverage. You know what a potent weapon this
is, and you know that this is Time Warner we're
talking about. This story on CNN is one of the first
exploratory feelers. Do people want to read about
another browser? Do people want to read about and
hear about alternatives? Well, some don't care, but
others do. All we're really waiting for here is a
slow news day. Sure, you cover something once and
nobody remembers it or cares. Cover it a few dozen
times and see what happens. They know what they're
doing.
* The ultimate weapon: version numbers. They've
pulled out the big 7.0 -- a step "ahead" of IE.
Version numbers don't matter? Well, not in terms
of actual quality they don't, but you just try to
convince end users that version numbers don't
matter. I tell you that Microsoft will be forced
to release IE 7 before they end of 2002, and will
face accusations in the IT community that it has
few improvements over IE 6. End users won't care
about these accusations, but it's a multi-front
war. To win, a browser has to win end users,
yes, but also IT people and web developers. IE
went nowhere, despite huge end-user adoption due
to bundling, until 5.0 came out and impressed the
IT crowd and the web developers. Until that,
the websites all still catered to NS4. All three
market segments matter. This puts Microsoft in
a tough position. IE6 already received lukewarm
praise from the IT people; it is barely better
than IE5.5, they say. But to keep end users
happy, IE7 _has_ to come out soon, because
Netscape is forcing their hand. That gives MS
precious little time to put together enough
improvements to avoid another lukewarm reception.
(They'll do it... this time. Even if they have
to buy their enhancements from NetCaptor. But
as I said the war is just beginning.)
Re:It's about flexibility
on
Gnome 2.0 RC1
·
· Score: 1
> Maybe/. hammered it and it needs some time...
No, but the computer was turned off for the weekend a few hours ago. It's a workstation, not a server per se, so it doesn't stay on all the time.
I looked at the logs before we closed shop, and it _did_ get hit more than I expected, but not enough to wipe out a T1. (It was, after all, nested a couple of levels into a thread, and the comment clearly stated that it was an older version of Gnome.)
It's about flexibility
on
Gnome 2.0 RC1
·
· Score: 1
> Looks like they are doing a good job in creating a Microsoft-like UI
Only with the default settings. Here's a screenshot showing (an older version of) Gnome looking a little different...
(That's on my workstation, so it'll become unreachable when I power down for the evening, sometime around 5:30pm EDT. Just as well; we're only on a T1 here...)
Linux is just a kernel. These systems presumably ship with a more complete Unix environment: shells, standard unix utils, an X server, and so on. It's correct to call them Unix-based.
It would also be correct to call them Linux-based, but that would be a different statement. Not that the target market will know the difference, but that's neither here nor there.
Unix does not mean BSD, per se. (Yes, it's trademarked, but IIRC not by anyone in Berkely.) It's a general category, more specific than POSIX, but not so specific as to designate that a particular kernel is used.
As I understand it, they charge for some applications and not others. Or something like that. Anyway, if the installs go smoothly enough, many end users wouldn't mind paying a small amount for the convenience. Those that do mind can always go download freely available software from the original distributors and take the trouble to read installation instructions... as I understand it, most unixy things should install as smoothly on Lindows as on any other Linux-kernel-based unix distribution. Installing Windows-based software under WINE would be trickier, and if the Lindows install process is smooth it could be a real benefit.
Of course, I haven't actually tested Lindows, so I have no idea how smooth its install process is in practice.
> Maybe sysadmins should know this, but Grandma User doesn't
Yes, I know this. That's why it's troubling. We know there will be droves and droves of unpatched systems out there.
I'm not sure what the solution is for that. Auto-installation of patches (as someone has pointed out XP does) has possibilities, but it has its own troublesome issues.
The real problem is that most users know nothing about security and care less. I don't think it's reasonable to expect that to change soon, either.
I also did not mean to completely exonerate Microsoft. They know and care more about security than a typical home user, but they leave something to be desired as well. Still, the problem would be difficult even if they cared enormously.
> About the crack about rednecks buying Lindows...what do you think > the average demographic of a Wally world online customer is? I'm > willing to bet it's not ma and pa kettle in podunk Arkansas.
College students and suburbanites mostly, I think...
But the point of the original redneck remark was that people who buy computers at Wall-Mart are not tech-savvy users, and that's true, as a general rule. Tech-savvy users buy their computers from small shops that build them, or they build them themselves, or they shop around. In any case they usually don't buy the true bargain-basement stuff, because they're planning to upgrade components as necessary and keep the thing running for several years, or if not it's because they can afford a new (nice) computer every year or so. These are power users.
People who buy computers at Wall-Mart are end users.
These are overgeneralisations, of course, but in general they are mostly true. Redneck is not the word I would have chosen, but the point made is valid. Think about droves of people buying these things who previously were not aware that Apple computers don't use Windows and had never heard of Linux, much less anything more obscure than that, and have no idea that Windows XP is based on NT ("huh?") rather than the consumer Windows line. Whether that makes them rednecks or just regular people in some field other than IT, the point is that they're not computer geeks. They're end users.
End user awareness that there are various operating systems to choose from is a good thing.
One of the fundamental principles of software develpment is that you can't find all the sticky problems until you get real users using the thing.
Consider Mozilla: progress was slow until the 0.9.x milestones, then all of a sudden it was good enough that a lot of users who tried it liked it enough to start using it as their regular browser, and whammo, the bugs started dropping like flies, and it shaped up incredibly in just a few weeks.
Same thing with Linux. Technical excellence aside, it was nowhere near ready for the typical end user until quite recently, but as the user base spreads beyond developers to end users, amazing strides are made in its _usability_ for end users. There's a breaking point somewhere, where enough users adopt a piece of software that the bugs show up and can be fixed. You don't reach that point without early adopters.
Assuming Windows 3.1 systems count, there
are certainly still more DOS systems out
there than Linux systems, even if you only
count systems connected to the internet.
However, these systems are mostly older
and probably don't have enough RAM to run
a modern web browser. I suspect most of
them are 486 class or less. DOS on modern
hardware is a fairly small niche market.
(I personally do keep a bootable DOS 6
partition around on my multiboot system,
but I don't use it anything like daily,
and I use Arachne for the web browser when
needed. And I don't claim to be typical.)
There was a DOSZilla project, but AFAIK
it is dead. I have heard nothing about
it for plus two years.
Slashdot Mirror
> Or as a compromise, set the firewalls (home and
> work) to only accept connections to port 22 from
> specified IP addresses
I suppose I could get into my home system from work
that way, but my home system is on dynamic IP, so I
still wouldn't be able to get into work from home,
if I go that route.
It does make sense. The math behind why has been pointed
out by others, so I'll just offer an analogy...
Is it faster to drive your car, or walk?
Depends how far you're going. If you're going several
miles, it's going to be a lot faster to drive. If you're
going across the street, it's faster to walk, because you
avoid the overhead of going out of your way to where the
car is parked, cleaning the snow off the windshield,
unlocking and starting the car, parking the car, and
getting out. When I was in high school, I walked. Once
a friend who was just arriving offered me a ride from the
end of my driveway. We ended up parking further from the
school door than where he picked me up.
The O(1) scheduler is the car. It'll get you to that
college in the next state faster, but it's no help for
getting across the street.
> One of the patches talkaout in the interview was supermount.
> Does anyone know why this is not in the main kernel.
Didn't know it wasn't... (Guess you know which distro I use.)
> This is a must have feature for linux on the desktop.
Agreed. _Especially_ for expansion into the non-geek
end-user segment of the desktop market (the largest
segment).
> It has been included in distros like mandrake for a long
> time, so it should be pretty stable.
It's been stable in my experience.
> The only version available is for Internet Explorer for Windows.
googlebar.mozdev.org
Unfortunately, this doesn't seem to install properly on
current versions of Mozilla, but it worked great with
0.9.9, and I suppose the bitrot will be fixed eventually.
Read/Write NTFS.
I would _use_ that. Currently to transfer things from
one OS to the other on a WinXP/Linux dual-boot system
is a pain (I'm using an smb share on a different PC on
the network...)
Yeah, I know WinXP can theoretically use FAT, but I
don't (call me crazy) particularly want to have to
reinstall it, and it came preinstalled on NTFS.
Three short snippets, and then commentary...
...
> His new role plays to perhaps his greatest skill--that
> uncanny ability to foresee how emerging software
> technologies can be woven together and parlayed into
> must-have "industry standard" products, which, in turn,
> reinforce demand for other software from Microsoft
> If Longhorn really does turn out to be a Super Windows--a
> big if--it will handle so many functions of computing that
> Oracle, Sun, AOL Time Warner, and Sony may find themselves
> with less to do...
> Because Gates' geeks are completely overhauling the
> operating system, they'll also have to redesign most of
> the company's other software products and services to take
> full advantage, including the MSN online service, its
> server applications,
If I read that right, we're saying three things:
1. Bill Gates' job is to interweave (i.e., make
interdependent) previously distinct technologies
so that Microsoft products are must-have.
2. Oracle, Sun, AOL, and Sony are targeted for
takeover. We already know the database filesystem
will make life tough for Oracle. So, what is the
threat to Sun, AOL, or Sony? I think I can answer
about AOL (below).
3. Read that third quote again. And again. Does
that sound to you like the MSN internet service
will have to be changed in order to be compatible
with the new MS OS? Does that mean, then, that
_only_ (the revised) MSN will work as an internet
service for users of the new MS OS?
Remember what was said in the Holloween Documents
about decommoditizing protocols? This is it; either
I've misread something badly, or Bill Gates wants to
"decommoditize" internet service. We already know
MSN users have to use Outlook; in a 2006, users
of this OS will have to use MSN (and, of course,
Outlook).
Fortunately, 2006 may be too late. In 1998 Unix
was not ready for the desktop. In 2002, Unix *is*
ready for the desktop. By 2006, I expect Microsoft
to have lost some of that market share to Linux
and OS X. With any luck, Longhorn will be too
late to lock down the whole market for them.
That's how I read it too. He's saying that if you install the
update and use privilege separation, you won't be vulnerable
during the time between when the bug and true fix are published
and the time you get them installed.
I'm thinking about going another route, and turning setting
things up so sshd only listens for connections from the local
network until after the bug and fix are published and I get
my systems patched. That loses me the ability to login to
my home system from work, or work systems from home, during
the interim, but it lets me wait for a stable fix.
Wireless networking is going to be a while yet
competing with wired networking. Except maybe
for line-of-sight wireless (which isn't really
the same thing; it may not have copper wires,
but it has lines the information has to follow
sure enough), wireless just isn't developed to
the point where it can offer as much bandwidth
as similarly priced non-wireless options. The
satellite stuff that's supposed to compete with
DSL for consumers costs twice as much per month
and has preposterous installation fees. AirPort
and similar local wireless options are dog slow
compared to even vanilla 10BaseT ethernet but
cost more like gigabit ethernet.
Basically, wireless just isn't _there_ yet.
Then there are security and privacy issues...
I don't think tech-savvy persons are more *or* less likely
to pirate music. I think the two traits are orthogonal.
Personally, I don't pirate music, but I also don't buy
this guy's stuff, because it's entirely out of genre for
me.
I almost never listen to 20th century music, except for the
occasional Yankovic or Shickele, or a little polyphonic
(non-monodic) a capella stuff, or stuff other people select
and play when I happen to be present. Mostly given the
choice I listen to baroque (especially late baroque), and
sometimes a little romantic (in the traditional sense;
Chopin or whatnot), or _occasionally_ some of the better
classical (I'm not into Mozart; Dvorak is pretty cool
though). I tire of monody (one lead melody part with
support from parts written around it) quickly and have a
marked preference for real polyphony (interwoven separate
but equal parts designed to go together) or fugue. So
as you can imagine I have no motivation to pirate anything
produced by Perl Jam.
Now, I'm not suggesting that tech savvy people are
unlikely to listen to modern music. What I am going
to suggest is that tech-savvy users have very specific
ideas about what music they like and will pay for, and
are less likely to buy an album just because it is
produced by a popular group, even a group that has
formerly produced albums they like. Part of what
makes a geek geeky is that he gets adamant about
small things other people don't seem to care about.
A geek will refuse to pay for something he does not
want, on principle, even if it's considered fashionable
outside of geek circles. (Unless it's a technical
gizmo he can mess with and reprogram, in which case
some geeks will crawl naked over a field of glass
caltrops to buy it, but nevermind; music does not
fit that category.) But I don't think geeks exhibit
a marked tendency to pirate, or not to pirate, any
more than the rest of the population at large.
Now, people who listen to baroque are probably less
likely to pirate music illegally than people who listen
to modern music, but that's a different matter. (Think
in terms of, lesse, 2002 less seventy is 1932... The
artist would have to be, err, 180 years old or so. Yes,
the performances are copyrighted, but the lack of composer
royalties drives the prices down a LOT. Plus, the
ecconomy of scale is quite significant for some of it.
Bach for example probably sells more albums every year
than this Moby guy has sold in his life (though perhaps
not more _dollars_ worth of albums).)
> The most fun part about a non-homebuilt machine is when
> you've got to get something proprietary replaced
Close, but the MOST fun is when you have to go to a
machine shop to get a custom screwdriver made.
HTH.HAND.
I also have had good experiences purchasing from tcwo.com
They ship promptly and were good about taking a personal
check (since I don't do credit cards) given adequate ID
(a driver's license number). All-around good experiences.
I've purchased from them several times.
Their selection is not exhaustive, but it is a good
representation of the things you are likely to need.
Most of the things I've needed that they didn't have
were specialty items that I ended up getting direct
from the relevant manufacturer (an Avant Stellar
keyboard from CVT, for example), although once when
I needed a lengthier cat5 patch cable I had to go
elsewhere (Computers4Sure IIRC) for it.
I had a different experience (one I would tend not
to charactarise as positive) attempting to purchase
from LinuxWarehouse, wherein I was sold an item that
was no longer in stock, and my order was cancelled
without notification, but my check was cashed...
I eventually got my refund, but it took months.
(Lest this be mistaken for libel, I can provide
specific details by email upon request, but I won't
bore all of slashdot with them.)
I think the etymology of the term will best answer your
question.
The original thinking was that there are not separate
plural and singular forms. This goes along with (part of) why
the term "spam" was chosen to represent the stuff. Like the
meat, there are no clear divisions between pieces; it's all
just _stuff_.
Think about it. If your mailbox contains 100 messages
(yeah, you wish it were so few...), and all of them are spam,
and 100 other people (again, never so few...) like you also
received 100 messages, all spam, and without sophisticated
IP subnet tracking none of you can easily tell who sent any
given message, and you're all receiving messages from a
different subset of the set of all spammers, and each spammer
sends you a different subset of his set of all messages...
So, any given message you received is very similar to a
random subset of the other ones you receved _and_ was also
received by a random subset of your 100 friends, each of
whom also has a random collection of messages that are
very similar to this one, but may not be exactly the
same as the ones you have...
What you have (all of you, collectively) is a messy
conglomeration of mixed parts. Like the Hormel product.
There is no important semantic distinction between one message
and the next, or between the same message received by different
people, or sent using different relays... it's not generally
useful to talk about an individual message. Collectively, it's
all just spam.
When you want to start filtering it, you (or your filtering
agent) have to look at each message individually. Especially
if you're building a database of known spam. But otherwise,
it's all just spam.
There were at one time some other terms to go along with
spam: jello and velveeta come to mind. The thinking was
that junk posts to usenet were different from junk email,
and that multiposts were different from crossposts, and
multiposts to one newsgroup repeatedly different from
(nearly) simultaneous multiposts to multiple groups, and
so on and so forth. These distinctions turned out to
be unimportant, largely because the same stuff gets sent
in all of the above ways, and so now it's all just spam.
I concur.
This is a good thing. Most people have very vague notions about what Windows and Microsoft even are; the advertising campaigns have some of them believing that it's an important part of computing, but only because all computers seem to have these buzzwords "Windows" and "Microsoft" attached to them. Exposure to the idea that computers don't need these buzzwords is a good thing.
In terms of user experience, users want to do a few simple things:
I'm glad Wall-Mart is no longer claiming that LindowsOS runs most MS programs. Lindows was not ready for that claim. But Linux *is* ready, or very close to ready, for the consumer desktop, as long as it comes preinstalled and preconfigured. I worry just a little about the silly-greeting-card thing... developers don't do such inane things, and I don't know whether anyone has put together a Print Artist equivalent for Linux.
First, I want to say that I don't consider dead tree copies
/. profile); that may not work as intended on other
to be useless in all cases. There are some books I'd
actually rather have in dead tree form -- mostly ones
that I'm never going to need to grep but will want to
take to the bathroom. For example, _Just for Fun_ (by
Torvalds), or Knuth's book on surreal numbers. Other
books (any reference book most programming books) I'd
rather have in electronic form.
But I really want to address the eye strain point.
Eye strain can be solved by changing your colour settings
and otherwise tweaking your environment.
Black on blinding white will of course cause eyestrain; even
most dead tree books that are meant to be read for hours on
end aren't printed on white paper, but a softer off-white
shade. More significantly, paper is an absorption/reflection
medium; it doesn't emit light. On a computer screen, you want
a darker background and a lighter foreground, if you want to
sit in front of it for days on end without eyestrain. Also,
a common mistake is to use the highest available contrast.
That's great for road signs, where you want to get the
reader's attention immediately, but for long exposure a
somewhat lower contrast is better. (Not _LOW_ contrast,
of course -- that's even worse than really high contrast.)
The best combination I've found so far is wheat (about
#F7DFB5; or so) on slate green (#305050;). Soft tertiary
colours are easy on the eyes. Yes, easier than paper,
if you have a good refresh rate on your monitor. Having
incandescent lighting (instead of fluorescent) helps too.
There's a Gnome/GTK theme, called thEmacs, that works
pretty well for me on my current (Mandrake distro,
Gnome session) system. For Windoze, you can do something
similar pretty easily, and I have. (For W95, you can get
my wheat on slate green and a couple of other colour
schemes from my website (click on "Personal", then on
"Colour Schemes I have Perpetrated" -- the base URL is
in my
versions of Windows, since gradients and things were
introduced in Windows 98 the format for colour schemes
in the registry may have changed).
The Inform Designer's Manual, by Graham Nelson.
4th edition if possible.
Dead tree copies of this one are slightly difficult
to find in stores, but this book is worth going out
of your way to get. It's not just loaded with
interesting examples of OO design in action -- it's
also one of the most _interesting_ programming
books ever written. The Inform language is niche,
but this book is worth reading for its own sake.
I'm sure it was so obvious you all just forgot to mention it,
that you all have it on your shelves both at work and at home,
but it really needs to be said anyway, in case there's some
poor newbie who just doesn't know...
You've gotta have the Hitchhiker's Guide to the Galaxy.
I'd also like to add that the Camel Book (Programming
Perl, already mentioned) is one of the best of the lot.
The browser war has only just barely begun. When MS first
announced IE, nobody took them seriously. Netscape bumbled
the first battle, and fell back to regroup. But the war is
about to begin in earnest soon. You can see the signs, if
you pay attention... AOL is getting ready to fight with
all the (considerable) weapons at their disposal. Including...
* Bundling. They're testing both IE and Gecko technology
for their next release. I posit that they're deciding
whether it's time, and what the response will be if they
switch. They've lost their AOL-on-desktop deal with
MS, so they _want_ to switch, but what will the user
response be? They're waiting to see.
Why do I think they're trying to decide whether it's
time? Because you can watch them guaging user response,
as they do tests with the new technology, distribute
it among a smaller user base (Compu$erve), and so on.
They're watching to see how it is received. Netscape
6 was received poorly, and they waited. Netscape 7,
now in beta, already has better response than 6. Don't
think they won't notice.
When will it be time? IMO, soon. Mozilla's user
experience shaped up considerably over the winter;
something happened: people who tried 0.9.7 and
0.9.8 liked it enough not just to use it themselves,
but to recommend it. When 0.9.9 came out, mozilla.org
had to get mirrors and more bandwidth, to handle the
increased demand. Sure, these are all the lunatic
fringe, people who will try a new technology before
it's really popular. But here's my point: Netscape
6 was branched before all of this, and Netscape 7
comes after. There has been a fundamental shift,
meanwhile, in how it is received. AOL is watching
the response, and they're going to see that the
response that comes back from Netscape 7 will be
good. It will be time. That's my prediction.
* Media coverage. You know what a potent weapon this
is, and you know that this is Time Warner we're
talking about. This story on CNN is one of the first
exploratory feelers. Do people want to read about
another browser? Do people want to read about and
hear about alternatives? Well, some don't care, but
others do. All we're really waiting for here is a
slow news day. Sure, you cover something once and
nobody remembers it or cares. Cover it a few dozen
times and see what happens. They know what they're
doing.
* The ultimate weapon: version numbers. They've
pulled out the big 7.0 -- a step "ahead" of IE.
Version numbers don't matter? Well, not in terms
of actual quality they don't, but you just try to
convince end users that version numbers don't
matter. I tell you that Microsoft will be forced
to release IE 7 before they end of 2002, and will
face accusations in the IT community that it has
few improvements over IE 6. End users won't care
about these accusations, but it's a multi-front
war. To win, a browser has to win end users,
yes, but also IT people and web developers. IE
went nowhere, despite huge end-user adoption due
to bundling, until 5.0 came out and impressed the
IT crowd and the web developers. Until that,
the websites all still catered to NS4. All three
market segments matter. This puts Microsoft in
a tough position. IE6 already received lukewarm
praise from the IT people; it is barely better
than IE5.5, they say. But to keep end users
happy, IE7 _has_ to come out soon, because
Netscape is forcing their hand. That gives MS
precious little time to put together enough
improvements to avoid another lukewarm reception.
(They'll do it... this time. Even if they have
to buy their enhancements from NetCaptor. But
as I said the war is just beginning.)
> Maybe /. hammered it and it needs some time...
No, but the computer was turned off for the weekend a few hours
ago. It's a workstation, not a server per se, so it doesn't stay
on all the time.
I looked at the logs before we closed shop, and it _did_ get
hit more than I expected, but not enough to wipe out a T1.
(It was, after all, nested a couple of levels into a thread,
and the comment clearly stated that it was an older version
of Gnome.)
> Looks like they are doing a good job in creating a Microsoft-like UI
1 54 824_shot.png
Only with the default settings. Here's a screenshot showing (an
older version of) Gnome looking a little different...
http://adminsystem.galion.lib.oh.us/2002_06_14_
(That's on my workstation, so it'll become unreachable when
I power down for the evening, sometime around 5:30pm EDT. Just
as well; we're only on a T1 here...)
Linux is just a kernel. These systems presumably ship with a
more complete Unix environment: shells, standard unix utils,
an X server, and so on. It's correct to call them Unix-based.
It would also be correct to call them Linux-based, but that
would be a different statement. Not that the target market
will know the difference, but that's neither here nor there.
Unix does not mean BSD, per se. (Yes, it's trademarked, but
IIRC not by anyone in Berkely.) It's a general category,
more specific than POSIX, but not so specific as to designate
that a particular kernel is used.
> are they charging after the first 3 downloads?
As I understand it, they charge for some applications
and not others. Or something like that. Anyway, if
the installs go smoothly enough, many end users wouldn't
mind paying a small amount for the convenience. Those
that do mind can always go download freely available
software from the original distributors and take the
trouble to read installation instructions... as I
understand it, most unixy things should install as
smoothly on Lindows as on any other Linux-kernel-based
unix distribution. Installing Windows-based software
under WINE would be trickier, and if the Lindows install
process is smooth it could be a real benefit.
Of course, I haven't actually tested Lindows, so I
have no idea how smooth its install process is in
practice.
> Maybe sysadmins should know this, but Grandma User doesn't
Yes, I know this. That's why it's troubling. We know there
will be droves and droves of unpatched systems out there.
I'm not sure what the solution is for that. Auto-installation
of patches (as someone has pointed out XP does) has possibilities,
but it has its own troublesome issues.
The real problem is that most users know nothing about
security and care less. I don't think it's reasonable to
expect that to change soon, either.
I also did not mean to completely exonerate Microsoft.
They know and care more about security than a typical
home user, but they leave something to be desired as
well. Still, the problem would be difficult even if
they cared enormously.
> About the crack about rednecks buying Lindows...what do you think
> the average demographic of a Wally world online customer is? I'm
> willing to bet it's not ma and pa kettle in podunk Arkansas.
College students and suburbanites mostly, I think...
But the point of the original redneck remark was that people
who buy computers at Wall-Mart are not tech-savvy users, and
that's true, as a general rule. Tech-savvy users buy their
computers from small shops that build them, or they build
them themselves, or they shop around. In any case they
usually don't buy the true bargain-basement stuff, because
they're planning to upgrade components as necessary and keep
the thing running for several years, or if not it's because
they can afford a new (nice) computer every year or so.
These are power users.
People who buy computers at Wall-Mart are end users.
These are overgeneralisations, of course, but in general
they are mostly true. Redneck is not the word I would
have chosen, but the point made is valid. Think about
droves of people buying these things who previously
were not aware that Apple computers don't use Windows
and had never heard of Linux, much less anything more
obscure than that, and have no idea that Windows XP
is based on NT ("huh?") rather than the consumer Windows
line. Whether that makes them rednecks or just regular
people in some field other than IT, the point is that
they're not computer geeks. They're end users.
End user awareness that there are various operating
systems to choose from is a good thing.
One of the fundamental principles of software
develpment is that you can't find all the
sticky problems until you get real users
using the thing.
Consider Mozilla: progress was slow until
the 0.9.x milestones, then all of a sudden
it was good enough that a lot of users who
tried it liked it enough to start using it
as their regular browser, and whammo, the
bugs started dropping like flies, and it
shaped up incredibly in just a few weeks.
Same thing with Linux. Technical excellence
aside, it was nowhere near ready for the
typical end user until quite recently, but
as the user base spreads beyond developers
to end users, amazing strides are made in
its _usability_ for end users. There's a
breaking point somewhere, where enough
users adopt a piece of software that the
bugs show up and can be fixed. You don't
reach that point without early adopters.
Assuming Windows 3.1 systems count, there are certainly still more DOS systems out there than Linux systems, even if you only count systems connected to the internet. However, these systems are mostly older and probably don't have enough RAM to run a modern web browser. I suspect most of them are 486 class or less. DOS on modern hardware is a fairly small niche market. (I personally do keep a bootable DOS 6 partition around on my multiboot system, but I don't use it anything like daily, and I use Arachne for the web browser when needed. And I don't claim to be typical.) There was a DOSZilla project, but AFAIK it is dead. I have heard nothing about it for plus two years.