> Don't know what's so magical about "1.0+" but it's certainly there.
Dunno. I generally won't *touch* a 1.0 release unless I know it's been through at least a year's worth of 0.x releases first. (For example, I had no qualms about Mozilla 1.0.) Frankly, I'm usually suspicious of anything.0, whether it's 1.0 or 12.0 -- I usually hold out for the.2 release. For example, I went from Mandrake 8.2 to 9.2 and am not touching 10.x until 10.2 comes out. This isn't a new trend either; DOS 5.0 and 6.0 both had serious bugs.
Some OSS projects buck this trend, doing conservative development and bug fixing on the old tree for a while before promoting it to the next X.0. Mozilla tends to work that way. But in general, anything.0 is anathematic.
No, not open source, but it's $0, easier to use than Outlook, more featureful, and approximately eleventy hillion jillion times more secure. (Not only does it not spread worms without user intervention, it roughly triples the number of steps required for a user to execute an executable attachment of any kind, and one of the steps involves a scary-looking warning dialog box with the word "Virus" in the title. I've seen people get trojans from Mozilla Mail, but there is not a single documented case of this ever happening to anyone with Pegasus Mail.) I have most of my non-geek family using it. It's great.
Personally, I use Gnus, but you can't tell most non-geeks to use that. Well, you technically *can* tell them to, but it's not a good idea.
> my time should be spent using my computer not fighting to update it
Yes, and we all know how easy it is to get the latest Firefox binary releases running on a Gnome1 system. All you need is GTK2. And the five or six libs that requires. And the things they require, including newer versions of some core C libraries that are incompatible with the rest of your system... why do you think I'm still using SeaMonkey/Navigator at work? At home I have the luxury of doing things like spending fifteen hours updating to a newer version of my distribution and getting that configured into a usable state, then fiddling endlessly for two weeks to *try* to get Gnome2 up to the standard of functionality of Gnome1, only to find out that's really totally impossible. At work, I really don't have time for that.
None of this is Mozilla.org's fault, mind you (except maybe the part about there being no prebuilt Firefox binaries for Gnome1), but it's a pain anyway.
No, you're talking out of the perspective of someone who works at a shelter, where you see 100% of the failures and a much, MUCH lower percentage of the successes. Your impression of the ratios is guaranteed to be skewed, by virtue of your situation.
It's like asking a marriage counsellor to estimate the percentage of marriages that are healthy. He only sees the ones that are falling apart, so he's going to give you a depressing answer, one that's a good deal worse than reality.
> And for every ten people who get a dog because they want some protection, > I'd say that at least seven of those dogs wind up in the shelter.
That's a gross exaggeration. You see every last one (in your area) that ends up in a shelter because some idiot doesn't really have the patience for a dog, but you never see the ones that actual human beings get and care for. The people who dump the dogs on the shelter (or in random places where so that the shelter ends up with them) are cut from the same cloth as the people who keep the dog chained in the back yard. People without feeling or a sense of responsibility. People who also use the television as a babysitter, like as not. Losers. Real people know how to take care of a dog, and we're talking here about a lonely old lady who will probably love the dog like a grandchild.
Like I said, the watchdog schtick is really an excuse for having the dog you really want to have around anyway.
> Is that American for GMT? Is that American for GMT?
No. It's French (or ISO maybe) for Universal Coordinated Time. For practical purposes it is functionally identical to GMT, however, and is defined that way deliberately, to avoid confusing everyone too much. The difference between GMT and UTC is highly technical and involves atomic clocks and leap seconds.
For intrusion detection, it's hard to beat a good dog. Get a medium-sized breed. Small yippie dogs like Chihuahuas bark at thin air all the time, and so you learn to ignore them, like the boy who cried wolf too often. Oh, and don't get purebred -- besides the extra cost, they have too many health issues. Get a medium-sized mutt with some collie or setter or shepherd heritage, or something. A shepherd/collie mix is good. Get it young, as soon as it's weaned preferably; it'll be easier to train it that way.
Grated, a dog is a significant expense, but it's also a really cool thing to have around, and has a lot more value than just intrusion detection. Really, intrusion detection is just an *excuse* to get what you ought to have anyway.
As an added security bonus, the dog will scare off many small-time intruders, especially kids. But the ones it won't scare off, it will detect usually, so for added peace of mind combine it with a cellphone. (The police will be a lot more likely to come if there's an intruder in your house when you call, as opposed to there having been one there earlier.)
The real value of the pet though is peace of mind. Once you've had a burglar get past an alarm system, you'll never really _trust_ an alarm system again. You feel safer with a dog in the house.
Resist the urge to get a cat. Yeah, they eat less. But the effect is just not the same. The dog may wimp out and hide from the burglar, but he's going to make a lot of noise doing it; the cat will detect the burglar, but *you* may not know it. With a dog, you will know.
I'm assuming here you're mainly worried about stuff happening when you're at home (though as I said the dog will scare off many small-time intruders too). If you're worried about protecting valuable assets, either keep them in a more secure location or put them where they'll be overlooked as worthless.
I can go one-up on that. My boss recently told me with a straight face that the fonts in two columns in a newsletter I was doing up were "the same" and needed to be "more different". (Nevermind the issue of whether it's a good thing to use different fonts for different columns; she's the boss, and doing up newsletters is only a small part of my job anyway.) The one was Verdana (a very clean look, definitely sans-serif) and the other was Georgia (about as strongly seriffed as you can get without into the realm of caligraphic script or illegible-at-normal-size hyperdecorative extremes). As near as I can determine, the only thing they have in common (besides foundry) is that they both look respectably good. In frustration I switched one of them to a half-baked barely-legible non-hinted pixelated-as-all-Redmond script font for one of the columns, and she thought it was great. Fortunately we managed to talk her down to Comic Sans MS, which is goofy looking and very informal but at least legible and more-or-less decent-looking.
> You don't know how many times I've sat in front of a user's nice LCD monitor
Users who like LCD monitors don't have good enough eyesight to even *discuss* the issue of what looks good. LCDs are getting better, but the color accuracy of even a midrange CRT is so much better, it's night and day.
Buy new computers *now*. Do not put the old computers in the new office. If you do, the new office will smell of its own accord in a few weeks, and then replacing the computers won't solve the problem.
Alternately: there are a handful of aromas known to mankind that are stronger than pigs. Ammonia, PVC cement, Ranch salad dressing, that sort of thing. Put some of *that* stuff in the office, and nobody will complain about the pig smell;-)
> Ask an American and a Japanese if they are "good in math". The Japanese > will typically say "no", the American will say "yes".
I'll call you on this one. A *very* small percentage of Americans will answer "yes" to that question, and almost all of those who do claim to be good at math majored in it in college and got better grades in their major than in other subjects.
The premise of the book was that an asteroid impact made it a much more tricky prospect. More than once, ISTR. But it eventually got (re)done and humanity developed to the point of space flight again yadda yadda and they all lived happily ever after, the end.
> However, this isn't a feature that is going to drive sales.
Bingo. That hits it on the head right there. WinFS is an interesting concept, but in practice it's not going to drive a lot of new sales or upgrades, so it gets pushed onto the back burner when there's something more pressing. Right now, one of the more pressing things, the one we know about, is security. There are probably other things too that contribute to pushing back WinFS, things that may or may not be publically talked about at all, and if the are we may not know exactly how much of MS's resources they are consuming.
If I had to speculate, I'd say getting the new and improved CLI into Longhorn is probably a much higher priority than WinFS, because it's more likely to drive upgrades, and also more likely to keep people from migrating away from MS, especially in server space. Heck, if it's really good, it could even drive migrations *to* Microsoft. Most of the innovation in shells in the *nix world lately has been non-mainstream stuff like eshell; bash and tsch and so on are, in terms of their core feature set, largely static. If MS can pull off some really compelling innovation in their new CLI, it would earn them a lot more street cred among system administrators than doing interesting but largely unimportant database-fs stuff that was already done by a company that then went bankrupt. Of course, I'm speculating when I guess that the CLI is one of the things that took priority over WinFS. But anyway, something did, and it's not hard to see why: WinFS is just not all that compelling.
> Once you get it working and get used to it, you would feel like losing one > hand without it.
Not really. We've already had the core features of WinFS in the BeOS fs, but when people migrate from the BeOS to Windows or Linux, the only filesystem feature they frequently wish they had back is journalling, which we now have. There are lots of things to miss about BeOS, perhaps the most obvious being supereasy installation, but the database-like attributes of the filesystem just don't rate. Most of the uses that the BeOS put that to turned out to be annoying stunts with more downside than upside. For example, the BeOS Person files, which were supposed to be how you kept an address book, were theoretically zero-size files, because all the data was stored in attributes. (They didn't take up zero space on disk; they just listed their size as zero.) The downside of this was that you had to be very careful about how you backed up these files and could not store them on other types of filesystems, or you'd lose data. The upside, if you can call it one, was that by tweaking which attributes were displayed for certain folders you could re-use the graphical file manager as an address-book viewer. Why you'd want to do that remains an unanswered question.
As for the notion that it simplified searching, that's actually backward. Compared to text-based formats it complicated searching considerably. To search for an address book entry you had to know _which_ attribute would contain the information you were searching for. For example, if you wanted to find email that you'd received on a given low-traffic mailing list, you had to know whether the message was sent directly To: the list or whether the list was in the Cc: or what, or you had to construct your search to compensate for your lack of knowledge via boolean reasoning. (Heaven help you in situations where you don't have any idea what the attribute is called that contains the information you're searching for.) With a mail-directory system like Gnus nnml, you just grep -r for the list address. It doesn't matter which header field it occurs in, because it's all part of the file.
The reason WinFS keeps getting pushed back is because it's a "cool" feature with purely theoretical (and strictly geek-oriented) value, and Microsoft _probably_ knows that. It'll make news headlines, but it won't sell upgrades.
> real stats from a large entertainment company website
Large entertainment company websites tend toward the low-end of userdom, in terms of tech aptitude. Not as extreme as MSN, but more biased than Yahoo. Of course, W3C is biased heavily in the other direction, toward geekdom. Stats from Google or Amazon would be more meaningful, but Google no longer shows their browser stats publically, and I don't think Amazon ever did, nor Ebay or any other of the sort of sites that I would expect to get an unbiased userbase consisting of geeks and nongeeks alike.
Nevertheless, while the W3C numbers don't give us an accurate count of the percentage of users using Mozilla.org browsers, they do show an increase, and that is probably reliable as far as it goes. I think it's safe to say based on this article that Firefox usage has increased rather significantly over the last several months. What the exact percentage is, we don't really know, but it's up from whatever it was before.
If I had to guess percentage, I'd say 15% is wildly high, but 3% may be a little low, perhaps.
There's also the small matter of the discrepancy between raw number of users and raw number of hits. It is a truism that the top ten percent of users (in terms of amount of internet usage) account for *way* more than ten percent of the total page hits. I strongly suspect that the average geek causes way more page hits than the average non-geek. Consequently, when I say that 3% may be a bit low, I mean in terms of the number of hits, not the number of users. I doubt anywhere near 3% of *users* are using Firefox yet, but the ones who are may be generating 4-5% of the page hits -- especially if you take that as the percentage of pages deliberately loaded by the user (i.e., not popups).
On a side note, my Dad recently asked me if his computer has Mozilla FoxFire [sic]. He was asking some non-techie people on a mostly-non-techie hobby forum about a certain browsing pattern related specifically to that forum website, and how to get around hitting the back button a whole lot of times and stuff, and someone told him to get Firefox for the tabbed browsing. I showed him how to use the tabbed browsing feature in the Netscape 7.0 browser that he already has, and that made him happy. (I'll upgrade them to Firefox eventually, but I'm waiting for 1.something probably. My mom doesn't like frequent upgrades. Well, she says she doesn't. I don't think she can really tell the difference, but whenever she knows I've been upgrading stuff on the computer she claims it confuses her.)
> If i were microsoft, i would take IE out of the OS
Oh, yeah, the OEMs would go for that. They wouldn't be at all concerned that their customers would buy a computer that wasn't internet-ready and be angry and return it. Nah. That internet thing, our customers don't need that... There's no way the OEMs would bundle a browser, such as a rebranded Gateway or Dell build of Firefox, if MSIE weren't included with the OEM OS. Nor would there be any danger that they'd defect to a different OS altogether, or use the threat of doing so to demand special OEM bundle pricing on IE.
> I personally believe they beat Netscape by having a better product.
That doesn't jive with the timeline. By the time MSIE was clearly better than Netscape (IE 5.0 at the *earliest*, arguably not until 5.5), the first browser war was so far over that a lot of people had forgotten it. The Netscape market share was under 5% by then.
MSIE market share exploded from roughly 0% to well past 50% during the IE3 era. Surely you aren't going to say that IE3 was a superior product? Even if you think IE4 was better than Netscape 4 (a quite dubious claim), the handwriting was on the wall already before IE4 came out. It was all about bundling.
However, bundling swings both ways. MS can bundle IE with the OS, but they can't control (directly) what ISPs put in the "connection kit" CDs that they send out to new (or prospective) customers. They've got AOL locked in pretty solid now, and of course MSN, but most ISPs are free agents here. If outfits like Earthlink start defecting to Firefox, that will have a significant impact on browser market share.
And then there's the other tactic -- getting computer gurus and techies to install your product on all the computers they support, because it's easier to support. We have yet to see really how effective that can be, but I'm betting that well past half of all computers get occasional maintenance from a geek of some kind (possibly just a teenage relative, but nevertheless...), so if a large pecentage of geeks decide it's worthwhile to install a certain browser on every system they service, that could have a significant impact too. I don't think that would ever push Mozilla market share into the kind of complete dominance that IE enjoys now, but it could push it well into the important-minority category perhaps.
Encyclopedias are nothing like authoritative. It's not hard to find dubious or even erroneous information in print encyclopedias, so I don't know why Wikipedia would be any different. Encyclopedias are by their very nature tertiary sources at best. You don't use them for in-depth research. That's not what they're for. Encyclopedias give you an introduction to a topic, a sweeping overview. They acquaint you with the basic themes and ideas of a topic so that you know what you're looking for when you do further research.
Some of the worst articles in Wikipedia come from the 1911 Encylopedia Britannica. For example, until a few months ago the Wikipedia article on Abraham came from the 1911 EB, and it was a horrible mishmash of irrelevant speculative non-NPOV drivel about the authorship of Genesis and sundry other nonsense, with almost nothing about Abraham's relevance to Judaism, Christianity, or Islam, and the information about the Genesis account was a very poor synopsis indeed, nitpicking fine points in some sections and glossing over major events and whole chapters in other places. The current article on Wikipedia is imperfect, of course, but *much* better.
Also, I really don't see how the "experiment" is remotely fair, inserting one error a day for five days and concluding the experiment on the sixth day. Print encyclopedias aren't expected to find their errors in under two weeks after the author first submits the article, are they?
I will readily concede that Wikipedia surely contains many errors and cannot be considered really authoritative, but I would say that goes with the territory of being an encyclopedia, and I don't see how it diminishes the value of Wikipedia as an encyclopedia, a tertiary source very useful for quickly finding a general overview of almost any topic.
> Some workplaces (like mine) have instituted a no Exploder policy.
It may not be necessary to go that far. I just removed the IE shortcuts from all the desktops and replaced them with shortcuts to other browsers (mostly Seamonkey/Navigator, because it's a mature product, but I've begun field testing Firefox and it seems to be close to ready; I should be able to deploy it fully in a couple more milestones, hopefully). I have not received a single complaint about this; MSIE is still available on staff workstations, in the Start menu under Applications, but they basically never use it, since it's so much more convenient to just hit the desktop icon for Mozilla. I did have to invest a small amount of training in making the staff aware that Mozilla was the recommended web browser. (I told them it was basically the same as Netscape, but a different brand. It probably helped that a number of them were previously familiar with Netscape 4.) I made the switch circa 2000, when Netscape 4 didn't see *quite* so preposterously out of date, though it was getting there. IE was at 5.0 at the time, which was really not *that* much better than NN4 (except at rendering stylesheets, which only a few sites used at the time). IE 5.5 was the version that really put IE cleanly ahead of NN4, and that wasn't out yet. Actually, many of the computers in the library still had Netscape 4, so Mozilla 0.9.x was a natural upgrade path for them -- and it crashed less often than Netscape 4, too. Then circa 0.9.5 I started the process of standardizing on it and removing the desktop shortcuts to the other browsers. I didn't actually uninstall, though; it wasn't necessary.
Re:Obligatory USian Viewpoint
on
Make Money Fast
·
· Score: 1
> At least it's harder to make Canadian money, with all the colours, > than the green USD.
With "modern" printing technology (i.e., since circa 1900), ink color is very easy to counterfeit, no matter which color of ink you use (assuming it's a static color, not irridescent or whatnot). The features that make money hard to counterfeit are other things, like the paper (still a biggie after all these years, the paper US money is printed on is fairly distinctive paper), the watermarks, and so on and so forth.
What's really clever is that different countermeasures are intended to defeat different sizes of counterfeitting operations. A small operation run by one or two people and creating a few hundred dollars a week in fake currency could get away (for a while, anyway) with ignoring some of the more subtle features, such as the security strips and the watermarks, because by the time the forgeries would be detected, the money would have changed hands half a dozen times already, and it would be hard to trace back its origin. But it'd be virtually impossible for such a small operation to get a decent approximation of the paper, and without that their counterfeit money would be detected the first place they tried to spend it. An operation large enough to be able to get closer on the paper would due to its size be easier to trace back to its source once the more subtle deviations (such as missing the watermark or the microprinting) were detected (at banks or wherever). Here I'm speaking mostly about US money, because that's what I'm familiar with, but I suspect other countries have similar systems, at least in the first world.
> I'm a heavy Photoshop user myself, and I prefer it - but mostly because of > the more polished tools like the masking, filters, and selection tools.
Photoshop's selection tools -- you *like* them? I've only used Photoshop a little bit, but my impression was that the selection tools were one of its greater weaknesses, when you stack it against Gimp[1]. The magic wand is less versatile, and select-by-color, unless I'm missing something, is just plain missing. Now, the version of Photoshop I was using was is one version old, so it's not fair to talk about Gimp2's superior path tool, since 1.3's Bezier tool wasn't so great. I suppose the current version of Photoshop has made some improvements. Like, I sure hope it's fixed the bug where you can't copy to the clipboard and paste into another application (even another app made by the same company, such as Pagemaker) without flattening the image first -- even if the image has only one layer.
But I didn't give up on Photoshop mostly because of Photoshop's shortcomings. I gave up on trying to do desktop publishing on the old iMac at work mostly because of Pagemaker's shortcomings, and switched to using OO.o on a PC, which doesn't have Photoshop. To date I know of two features Pagemaker has that OO.o doesn't have; one of them I never use[2], and the other[3] is a UI convenience issue, and there's an issue for it in issuezilla. That's pretty impressive, considering OO.o is actually just a glorified word processor that doesn't really bill itself as a desktop publishing app. But you can do quite a lot of stuff with its frames and tables. It was the complete and utter lack of any sensible or convenient way to do tables in Pagemaker that broke the camel's back and occasioned my switch.
I experimented briefly with Scribus, but it has an accessibility bug[4] most people will never notice that makes it unusable for me.
[1] That and the inferior way it handles layers, and the inferior stock
collection of filters, though I've heard there are add-on packs you
can get to solve the latter problem.
[2] The ability to (over)flow text from one frame into another. It would
be useful for a newspaper to do continued-on-page-A4 type of stuff, but
for smaller publications I've found most[5] of the cases where people
use it it's an ugly kludge for what ought to be done with anchoring and
wrapping properties, except that Pagemaker won't let you e.g. put an
image inside a text frame and wrap the text around it, so you have to
do things the hard way.
[3] Easy way to rotate objects at angles other than (multiples of) 90 degrees.
You can do it, but you have to put the object into a Draw document, then
bring up the Format Position and Size dialog, enter an angle... it's a
royal pain. This is something Pagemaker gets right.
[4] It doesn't adhere to system on-screen color settings, instead forcing a
blinding white background on the user. OO.o for Linux just finally
fixed this in 1.1, though StarOffice always did it right on Windows,
even in the ancient Kruftier-than-Kruftitude-Itself days of prehistoric
StarDivision yore.)
[4] Notice I did not claim "all". I'm sure if you've done much desktop
publishing you can think of an instance where you'd have a legitimate
want for this feature. I can *think* of such instances also, but in
practice I don't generally run into those situations much.
> You go over to a friends house that has broadband and a CD-Writer (both very > popular these days) and download the patches onto a CD-R and take it home?
Huh? This makes no sense. I read/., which means I'm probably a computer geek. Among other things, this means that if good broadband were *available* in my area, I'd have it. It's not, so I don't. Also, it means that most of my friends live hundreds of miles away, and the only regular contact I have with them is via the internet. "go over to a friend's house" is a nonsensical statement; by the time I could plan enough time off work and arrange for the trip, I could just use wget to download half a dozen ISO images over my dialup connection.
Mac is way more different from Windows than other *nix systems are. You're going to find thousands of little things you never imagined would be different. Taken individually, none of them are a big deal.
As for the mouse, spend the ten bucks and get yourself a USB scrollmouse. Most major Mac software knows how to use the extra buttons (including a middle button) if you have them, so if you're accustomed to a multibutton mouse, get one. One less difference to get used to.
The things that will bug you are much more subtle differences... * You switch between windows on Windows, but on a Mac you switch between apps. * On Windows if you close an app's last window, it exits; on a Mac, it doesn't.
(This was really annoying on classic versions, because doubleclicking the
app's desktop shortcut subsequently didn't open a window. OS X fixes this.) * You can only resize a window from the lower-right-hand corner. * There is no maximize. There's a "resize", but it's nothing like the same.
(On the plus side, you get windowshading, which Windows lacks.) * The filesystem is arranged rather differently. (I don't mean the lowlevel
technical details, but rather where different types of files are kept.)
Everything you know about where things are on the hard drive -- forget it. * Menubars work rather differently.
I'm really just scratching the surface here. Everything is different, from how you open zip files to the fundamental dissimilarity of the dock to the Windows taskbar. There's a reason Mac people don't like Windows: it's nothing like what they're used to.
Slashdot Mirror
> Don't know what's so magical about "1.0+" but it's certainly there.
.0, whether .2 release. For example, I
Dunno. I generally won't *touch* a 1.0 release unless I know it's been through
at least a year's worth of 0.x releases first. (For example, I had no qualms
about Mozilla 1.0.) Frankly, I'm usually suspicious of anything
it's 1.0 or 12.0 -- I usually hold out for the
went from Mandrake 8.2 to 9.2 and am not touching 10.x until 10.2 comes out.
This isn't a new trend either; DOS 5.0 and 6.0 both had serious bugs.
Some OSS projects buck this trend, doing conservative development and bug
fixing on the old tree for a while before promoting it to the next X.0.
Mozilla tends to work that way. But in general, anything.0 is anathematic.
What are the winmail.dat files, and what is their purpose, and what should
the mailreader do with them? I don't recall ever seeing one...
> Thunderbird still needs to catch up to Outlook.
Two words: Pegasus Mail.
No, not open source, but it's $0, easier to use than Outlook, more featureful,
and approximately eleventy hillion jillion times more secure. (Not only does
it not spread worms without user intervention, it roughly triples the number of
steps required for a user to execute an executable attachment of any kind,
and one of the steps involves a scary-looking warning dialog box with the
word "Virus" in the title. I've seen people get trojans from Mozilla Mail,
but there is not a single documented case of this ever happening to anyone
with Pegasus Mail.) I have most of my non-geek family using it. It's great.
Personally, I use Gnus, but you can't tell most non-geeks to use that.
Well, you technically *can* tell them to, but it's not a good idea.
> my time should be spent using my computer not fighting to update it
Yes, and we all know how easy it is to get the latest Firefox binary releases
running on a Gnome1 system. All you need is GTK2. And the five or six libs
that requires. And the things they require, including newer versions of some
core C libraries that are incompatible with the rest of your system... why do
you think I'm still using SeaMonkey/Navigator at work? At home I have the
luxury of doing things like spending fifteen hours updating to a newer version
of my distribution and getting that configured into a usable state, then
fiddling endlessly for two weeks to *try* to get Gnome2 up to the standard
of functionality of Gnome1, only to find out that's really totally impossible.
At work, I really don't have time for that.
None of this is Mozilla.org's fault, mind you (except maybe the part about
there being no prebuilt Firefox binaries for Gnome1), but it's a pain anyway.
> I'm not talking out of my butt here.
No, you're talking out of the perspective of someone who works at a shelter,
where you see 100% of the failures and a much, MUCH lower percentage of the
successes. Your impression of the ratios is guaranteed to be skewed, by
virtue of your situation.
It's like asking a marriage counsellor to estimate the percentage of marriages
that are healthy. He only sees the ones that are falling apart, so he's going
to give you a depressing answer, one that's a good deal worse than reality.
> And for every ten people who get a dog because they want some protection,
> I'd say that at least seven of those dogs wind up in the shelter.
That's a gross exaggeration. You see every last one (in your area) that ends
up in a shelter because some idiot doesn't really have the patience for a dog,
but you never see the ones that actual human beings get and care for. The
people who dump the dogs on the shelter (or in random places where so that
the shelter ends up with them) are cut from the same cloth as the people who
keep the dog chained in the back yard. People without feeling or a sense of
responsibility. People who also use the television as a babysitter, like as
not. Losers. Real people know how to take care of a dog, and we're talking
here about a lonely old lady who will probably love the dog like a grandchild.
Like I said, the watchdog schtick is really an excuse for having the dog you
really want to have around anyway.
> Is that American for GMT? Is that American for GMT?
No. It's French (or ISO maybe) for Universal Coordinated Time. For practical
purposes it is functionally identical to GMT, however, and is defined that way
deliberately, to avoid confusing everyone too much. The difference between GMT
and UTC is highly technical and involves atomic clocks and leap seconds.
For intrusion detection, it's hard to beat a good dog. Get a medium-sized
breed. Small yippie dogs like Chihuahuas bark at thin air all the time, and
so you learn to ignore them, like the boy who cried wolf too often. Oh, and
don't get purebred -- besides the extra cost, they have too many health issues.
Get a medium-sized mutt with some collie or setter or shepherd heritage, or
something. A shepherd/collie mix is good. Get it young, as soon as it's
weaned preferably; it'll be easier to train it that way.
Grated, a dog is a significant expense, but it's also a really cool thing to
have around, and has a lot more value than just intrusion detection. Really,
intrusion detection is just an *excuse* to get what you ought to have anyway.
As an added security bonus, the dog will scare off many small-time intruders,
especially kids. But the ones it won't scare off, it will detect usually,
so for added peace of mind combine it with a cellphone. (The police will be
a lot more likely to come if there's an intruder in your house when you call,
as opposed to there having been one there earlier.)
The real value of the pet though is peace of mind. Once you've had a burglar
get past an alarm system, you'll never really _trust_ an alarm system again.
You feel safer with a dog in the house.
Resist the urge to get a cat. Yeah, they eat less. But the effect is just
not the same. The dog may wimp out and hide from the burglar, but he's going
to make a lot of noise doing it; the cat will detect the burglar, but *you*
may not know it. With a dog, you will know.
I'm assuming here you're mainly worried about stuff happening when you're at
home (though as I said the dog will scare off many small-time intruders too).
If you're worried about protecting valuable assets, either keep them in a
more secure location or put them where they'll be overlooked as worthless.
I can go one-up on that. My boss recently told me with a straight face that
the fonts in two columns in a newsletter I was doing up were "the same" and
needed to be "more different". (Nevermind the issue of whether it's a good
thing to use different fonts for different columns; she's the boss, and doing
up newsletters is only a small part of my job anyway.) The one was Verdana
(a very clean look, definitely sans-serif) and the other was Georgia (about
as strongly seriffed as you can get without into the realm of caligraphic
script or illegible-at-normal-size hyperdecorative extremes). As near as I
can determine, the only thing they have in common (besides foundry) is that
they both look respectably good. In frustration I switched one of them to a
half-baked barely-legible non-hinted pixelated-as-all-Redmond script font
for one of the columns, and she thought it was great. Fortunately we managed
to talk her down to Comic Sans MS, which is goofy looking and very informal
but at least legible and more-or-less decent-looking.
> You don't know how many times I've sat in front of a user's nice LCD monitor
Users who like LCD monitors don't have good enough eyesight to even *discuss*
the issue of what looks good. LCDs are getting better, but the color accuracy
of even a midrange CRT is so much better, it's night and day.
Buy new computers *now*. Do not put the old computers in the new office. If
;-)
you do, the new office will smell of its own accord in a few weeks, and then
replacing the computers won't solve the problem.
Alternately: there are a handful of aromas known to mankind that are stronger
than pigs. Ammonia, PVC cement, Ranch salad dressing, that sort of thing.
Put some of *that* stuff in the office, and nobody will complain about the
pig smell
> Ask an American and a Japanese if they are "good in math". The Japanese
> will typically say "no", the American will say "yes".
I'll call you on this one. A *very* small percentage of Americans will
answer "yes" to that question, and almost all of those who do claim to be
good at math majored in it in college and got better grades in their major
than in other subjects.
The premise of the book was that an asteroid impact made it a much more
tricky prospect. More than once, ISTR. But it eventually got (re)done
and humanity developed to the point of space flight again yadda yadda and
they all lived happily ever after, the end.
> However, this isn't a feature that is going to drive sales.
Bingo. That hits it on the head right there. WinFS is an interesting concept,
but in practice it's not going to drive a lot of new sales or upgrades, so it
gets pushed onto the back burner when there's something more pressing. Right
now, one of the more pressing things, the one we know about, is security.
There are probably other things too that contribute to pushing back WinFS,
things that may or may not be publically talked about at all, and if the are
we may not know exactly how much of MS's resources they are consuming.
If I had to speculate, I'd say getting the new and improved CLI into Longhorn
is probably a much higher priority than WinFS, because it's more likely to
drive upgrades, and also more likely to keep people from migrating away from
MS, especially in server space. Heck, if it's really good, it could even drive
migrations *to* Microsoft. Most of the innovation in shells in the *nix world
lately has been non-mainstream stuff like eshell; bash and tsch and so on are,
in terms of their core feature set, largely static. If MS can pull off some
really compelling innovation in their new CLI, it would earn them a lot more
street cred among system administrators than doing interesting but largely
unimportant database-fs stuff that was already done by a company that then
went bankrupt. Of course, I'm speculating when I guess that the CLI is one
of the things that took priority over WinFS. But anyway, something did, and
it's not hard to see why: WinFS is just not all that compelling.
> Once you get it working and get used to it, you would feel like losing one
> hand without it.
Not really. We've already had the core features of WinFS in the BeOS fs, but
when people migrate from the BeOS to Windows or Linux, the only filesystem
feature they frequently wish they had back is journalling, which we now have.
There are lots of things to miss about BeOS, perhaps the most obvious being
supereasy installation, but the database-like attributes of the filesystem
just don't rate. Most of the uses that the BeOS put that to turned out to
be annoying stunts with more downside than upside. For example, the BeOS
Person files, which were supposed to be how you kept an address book, were
theoretically zero-size files, because all the data was stored in attributes.
(They didn't take up zero space on disk; they just listed their size as zero.)
The downside of this was that you had to be very careful about how you backed
up these files and could not store them on other types of filesystems, or
you'd lose data. The upside, if you can call it one, was that by tweaking
which attributes were displayed for certain folders you could re-use the
graphical file manager as an address-book viewer. Why you'd want to do that
remains an unanswered question.
As for the notion that it simplified searching, that's actually backward.
Compared to text-based formats it complicated searching considerably. To
search for an address book entry you had to know _which_ attribute would
contain the information you were searching for. For example, if you wanted
to find email that you'd received on a given low-traffic mailing list, you
had to know whether the message was sent directly To: the list or whether
the list was in the Cc: or what, or you had to construct your search to
compensate for your lack of knowledge via boolean reasoning. (Heaven help
you in situations where you don't have any idea what the attribute is called
that contains the information you're searching for.) With a mail-directory
system like Gnus nnml, you just grep -r for the list address. It doesn't
matter which header field it occurs in, because it's all part of the file.
The reason WinFS keeps getting pushed back is because it's a "cool" feature
with purely theoretical (and strictly geek-oriented) value, and Microsoft
_probably_ knows that. It'll make news headlines, but it won't sell upgrades.
From the Sci-Fi book _Terraforming Earth_, which is based on this premise.
> real stats from a large entertainment company website
Large entertainment company websites tend toward the low-end of userdom, in
terms of tech aptitude. Not as extreme as MSN, but more biased than Yahoo.
Of course, W3C is biased heavily in the other direction, toward geekdom.
Stats from Google or Amazon would be more meaningful, but Google no longer
shows their browser stats publically, and I don't think Amazon ever did, nor
Ebay or any other of the sort of sites that I would expect to get an unbiased
userbase consisting of geeks and nongeeks alike.
Nevertheless, while the W3C numbers don't give us an accurate count of the
percentage of users using Mozilla.org browsers, they do show an increase,
and that is probably reliable as far as it goes. I think it's safe to say
based on this article that Firefox usage has increased rather significantly
over the last several months. What the exact percentage is, we don't really
know, but it's up from whatever it was before.
If I had to guess percentage, I'd say 15% is wildly high, but 3% may be a
little low, perhaps.
There's also the small matter of the discrepancy between raw number of users
and raw number of hits. It is a truism that the top ten percent of users
(in terms of amount of internet usage) account for *way* more than ten
percent of the total page hits. I strongly suspect that the average geek
causes way more page hits than the average non-geek. Consequently, when I
say that 3% may be a bit low, I mean in terms of the number of hits, not the
number of users. I doubt anywhere near 3% of *users* are using Firefox yet,
but the ones who are may be generating 4-5% of the page hits -- especially
if you take that as the percentage of pages deliberately loaded by the user
(i.e., not popups).
On a side note, my Dad recently asked me if his computer has Mozilla FoxFire
[sic]. He was asking some non-techie people on a mostly-non-techie hobby
forum about a certain browsing pattern related specifically to that forum
website, and how to get around hitting the back button a whole lot of times
and stuff, and someone told him to get Firefox for the tabbed browsing. I
showed him how to use the tabbed browsing feature in the Netscape 7.0 browser
that he already has, and that made him happy. (I'll upgrade them to Firefox
eventually, but I'm waiting for 1.something probably. My mom doesn't like
frequent upgrades. Well, she says she doesn't. I don't think she can really
tell the difference, but whenever she knows I've been upgrading stuff on the
computer she claims it confuses her.)
> If i were microsoft, i would take IE out of the OS
Oh, yeah, the OEMs would go for that. They wouldn't be at all concerned that
their customers would buy a computer that wasn't internet-ready and be angry
and return it. Nah. That internet thing, our customers don't need that...
There's no way the OEMs would bundle a browser, such as a rebranded Gateway
or Dell build of Firefox, if MSIE weren't included with the OEM OS. Nor
would there be any danger that they'd defect to a different OS altogether,
or use the threat of doing so to demand special OEM bundle pricing on IE.
> I personally believe they beat Netscape by having a better product.
That doesn't jive with the timeline. By the time MSIE was clearly better
than Netscape (IE 5.0 at the *earliest*, arguably not until 5.5), the first
browser war was so far over that a lot of people had forgotten it. The
Netscape market share was under 5% by then.
MSIE market share exploded from roughly 0% to well past 50% during the IE3
era. Surely you aren't going to say that IE3 was a superior product? Even
if you think IE4 was better than Netscape 4 (a quite dubious claim), the
handwriting was on the wall already before IE4 came out. It was all about
bundling.
However, bundling swings both ways. MS can bundle IE with the OS, but they
can't control (directly) what ISPs put in the "connection kit" CDs that they
send out to new (or prospective) customers. They've got AOL locked in pretty
solid now, and of course MSN, but most ISPs are free agents here. If outfits
like Earthlink start defecting to Firefox, that will have a significant impact
on browser market share.
And then there's the other tactic -- getting computer gurus and techies to
install your product on all the computers they support, because it's easier
to support. We have yet to see really how effective that can be, but I'm
betting that well past half of all computers get occasional maintenance from
a geek of some kind (possibly just a teenage relative, but nevertheless...),
so if a large pecentage of geeks decide it's worthwhile to install a certain
browser on every system they service, that could have a significant impact
too. I don't think that would ever push Mozilla market share into the kind
of complete dominance that IE enjoys now, but it could push it well into
the important-minority category perhaps.
Encyclopedias are nothing like authoritative. It's not hard to find dubious
or even erroneous information in print encyclopedias, so I don't know why
Wikipedia would be any different. Encyclopedias are by their very nature
tertiary sources at best. You don't use them for in-depth research. That's
not what they're for. Encyclopedias give you an introduction to a topic, a
sweeping overview. They acquaint you with the basic themes and ideas of a
topic so that you know what you're looking for when you do further research.
Some of the worst articles in Wikipedia come from the 1911 Encylopedia
Britannica. For example, until a few months ago the Wikipedia article on
Abraham came from the 1911 EB, and it was a horrible mishmash of irrelevant
speculative non-NPOV drivel about the authorship of Genesis and sundry other
nonsense, with almost nothing about Abraham's relevance to Judaism,
Christianity, or Islam, and the information about the Genesis account was a
very poor synopsis indeed, nitpicking fine points in some sections and
glossing over major events and whole chapters in other places. The current
article on Wikipedia is imperfect, of course, but *much* better.
Also, I really don't see how the "experiment" is remotely fair, inserting
one error a day for five days and concluding the experiment on the sixth day.
Print encyclopedias aren't expected to find their errors in under two weeks
after the author first submits the article, are they?
I will readily concede that Wikipedia surely contains many errors and cannot
be considered really authoritative, but I would say that goes with the
territory of being an encyclopedia, and I don't see how it diminishes the
value of Wikipedia as an encyclopedia, a tertiary source very useful for
quickly finding a general overview of almost any topic.
> Some workplaces (like mine) have instituted a no Exploder policy.
It may not be necessary to go that far. I just removed the IE shortcuts from
all the desktops and replaced them with shortcuts to other browsers (mostly
Seamonkey/Navigator, because it's a mature product, but I've begun field
testing Firefox and it seems to be close to ready; I should be able to deploy
it fully in a couple more milestones, hopefully). I have not received a
single complaint about this; MSIE is still available on staff workstations,
in the Start menu under Applications, but they basically never use it, since
it's so much more convenient to just hit the desktop icon for Mozilla. I did
have to invest a small amount of training in making the staff aware that
Mozilla was the recommended web browser. (I told them it was basically the
same as Netscape, but a different brand. It probably helped that a number of
them were previously familiar with Netscape 4.) I made the switch circa 2000,
when Netscape 4 didn't see *quite* so preposterously out of date, though it
was getting there. IE was at 5.0 at the time, which was really not *that*
much better than NN4 (except at rendering stylesheets, which only a few sites
used at the time). IE 5.5 was the version that really put IE cleanly ahead of
NN4, and that wasn't out yet. Actually, many of the computers in the library
still had Netscape 4, so Mozilla 0.9.x was a natural upgrade path for them --
and it crashed less often than Netscape 4, too. Then circa 0.9.5 I started
the process of standardizing on it and removing the desktop shortcuts to the
other browsers. I didn't actually uninstall, though; it wasn't necessary.
> At least it's harder to make Canadian money, with all the colours,
> than the green USD.
With "modern" printing technology (i.e., since circa 1900), ink color is very
easy to counterfeit, no matter which color of ink you use (assuming it's a
static color, not irridescent or whatnot). The features that make money hard
to counterfeit are other things, like the paper (still a biggie after all
these years, the paper US money is printed on is fairly distinctive paper),
the watermarks, and so on and so forth.
What's really clever is that different countermeasures are intended to defeat
different sizes of counterfeitting operations. A small operation run by one
or two people and creating a few hundred dollars a week in fake currency could
get away (for a while, anyway) with ignoring some of the more subtle features,
such as the security strips and the watermarks, because by the time the
forgeries would be detected, the money would have changed hands half a dozen
times already, and it would be hard to trace back its origin. But it'd be
virtually impossible for such a small operation to get a decent approximation
of the paper, and without that their counterfeit money would be detected the
first place they tried to spend it. An operation large enough to be able to
get closer on the paper would due to its size be easier to trace back to its
source once the more subtle deviations (such as missing the watermark or the
microprinting) were detected (at banks or wherever). Here I'm speaking mostly
about US money, because that's what I'm familiar with, but I suspect other
countries have similar systems, at least in the first world.
> I'm a heavy Photoshop user myself, and I prefer it - but mostly because of
> the more polished tools like the masking, filters, and selection tools.
Photoshop's selection tools -- you *like* them? I've only used Photoshop
a little bit, but my impression was that the selection tools were one of
its greater weaknesses, when you stack it against Gimp[1]. The magic wand
is less versatile, and select-by-color, unless I'm missing something, is just
plain missing. Now, the version of Photoshop I was using was is one version
old, so it's not fair to talk about Gimp2's superior path tool, since 1.3's
Bezier tool wasn't so great. I suppose the current version of Photoshop has
made some improvements. Like, I sure hope it's fixed the bug where you can't
copy to the clipboard and paste into another application (even another app
made by the same company, such as Pagemaker) without flattening the image
first -- even if the image has only one layer.
But I didn't give up on Photoshop mostly because of Photoshop's shortcomings.
I gave up on trying to do desktop publishing on the old iMac at work mostly
because of Pagemaker's shortcomings, and switched to using OO.o on a PC,
which doesn't have Photoshop. To date I know of two features Pagemaker has
that OO.o doesn't have; one of them I never use[2], and the other[3] is a UI
convenience issue, and there's an issue for it in issuezilla. That's pretty
impressive, considering OO.o is actually just a glorified word processor that
doesn't really bill itself as a desktop publishing app. But you can do quite
a lot of stuff with its frames and tables. It was the complete and utter
lack of any sensible or convenient way to do tables in Pagemaker that broke
the camel's back and occasioned my switch.
I experimented briefly with Scribus, but it has an accessibility bug[4] most
people will never notice that makes it unusable for me.
[1] That and the inferior way it handles layers, and the inferior stock
collection of filters, though I've heard there are add-on packs you
can get to solve the latter problem.
[2] The ability to (over)flow text from one frame into another. It would
be useful for a newspaper to do continued-on-page-A4 type of stuff, but
for smaller publications I've found most[5] of the cases where people
use it it's an ugly kludge for what ought to be done with anchoring and
wrapping properties, except that Pagemaker won't let you e.g. put an
image inside a text frame and wrap the text around it, so you have to
do things the hard way.
[3] Easy way to rotate objects at angles other than (multiples of) 90 degrees.
You can do it, but you have to put the object into a Draw document, then
bring up the Format Position and Size dialog, enter an angle... it's a
royal pain. This is something Pagemaker gets right.
[4] It doesn't adhere to system on-screen color settings, instead forcing a
blinding white background on the user. OO.o for Linux just finally
fixed this in 1.1, though StarOffice always did it right on Windows,
even in the ancient Kruftier-than-Kruftitude-Itself days of prehistoric
StarDivision yore.)
[4] Notice I did not claim "all". I'm sure if you've done much desktop
publishing you can think of an instance where you'd have a legitimate
want for this feature. I can *think* of such instances also, but in
practice I don't generally run into those situations much.
> You go over to a friends house that has broadband and a CD-Writer (both very
/., which means I'm probably a computer
> popular these days) and download the patches onto a CD-R and take it home?
Huh? This makes no sense. I read
geek. Among other things, this means that if good broadband were *available*
in my area, I'd have it. It's not, so I don't. Also, it means that most of my
friends live hundreds of miles away, and the only regular contact I have with
them is via the internet. "go over to a friend's house" is a nonsensical
statement; by the time I could plan enough time off work and arrange for the
trip, I could just use wget to download half a dozen ISO images over my dialup
connection.
VMS has versioning built into the filesystem. This is high on my list of
useful non-mainstream features that I'd like to see in more OSes.
Mac is way more different from Windows than other *nix systems are. You're going
to find thousands of little things you never imagined would be different. Taken
individually, none of them are a big deal.
As for the mouse, spend the ten bucks and get yourself a USB scrollmouse. Most
major Mac software knows how to use the extra buttons (including a middle button)
if you have them, so if you're accustomed to a multibutton mouse, get one. One
less difference to get used to.
The things that will bug you are much more subtle differences...
* You switch between windows on Windows, but on a Mac you switch between apps.
* On Windows if you close an app's last window, it exits; on a Mac, it doesn't.
(This was really annoying on classic versions, because doubleclicking the
app's desktop shortcut subsequently didn't open a window. OS X fixes this.)
* You can only resize a window from the lower-right-hand corner.
* There is no maximize. There's a "resize", but it's nothing like the same.
(On the plus side, you get windowshading, which Windows lacks.)
* The filesystem is arranged rather differently. (I don't mean the lowlevel
technical details, but rather where different types of files are kept.)
Everything you know about where things are on the hard drive -- forget it.
* Menubars work rather differently.
I'm really just scratching the surface here. Everything is different, from
how you open zip files to the fundamental dissimilarity of the dock to the
Windows taskbar. There's a reason Mac people don't like Windows: it's
nothing like what they're used to.