So, let me try to understand your vision: Players are running around shooting each other in real time, but someone is engaged in a bullet-time battle, they are shown as being very slow. However, this won't affect the rest of the world?
Anyone who can see into the bubble is part of the bubble, was what I meant. That's why I mentioned the rain; you have to constrain line-of-sight so the effect doesn't spread to the whole MMO world.
Also, what if you're in a team? You have five characters (Team A) who encounter 5 other characters (Team B). Two opposing characters, one from each team, go into bullet-time.
The players are all involved in the bubble. The players without bullet time enabled have slowed controls and move like they're stuck in mud. The characters with bullet time enabled move normally except that they appear less affected by gravity (9.8 ms-2 appears as 0.98ms-2 if one 'real' second is 0.1 'in-game' seconds)
Alternatively, while the two players are in bullet time, you take your four team members and spray as much machine-gun fire into the bubble as possible...
etc, etc - you've misunderstood. All players within sight of each other are affected by the bubble.
your control is stopped immediately when someone pops into bullet time, then you've just given controllable lag to upper-level characters
Yes, precisely (except that control is slowed, not stopped). What do you think bullet time is, but the ability for agents to avoid the lag of the real world? However, if you have any ability available, turning on your ability should be either automatic or instantaneous, not delayed as for normal motion.
Then you have to determine what the affected radius of bullet time exposure is, whether it works through walls and windows, etc.
You can have it purely line of sight, but that has obvious problems with 'infecting' the whole world with slowness, unless you prevent people forming a human chain of lines of sight to the original bubble.
My preferred alternative would be to make the effect depend on your radial distance from a character in bullet time. In this case, yes, bullets from outside would slow as they neared the location of characters in bullet time; but if the radius of effect is large enough - say linearly decreasing to 100m - you still have plenty of time to see bullets coming.
Putting some real figures on this - 500m/s muzzle velocity is realistic, 5m/s is dodgeable so 1/100 speed is agent-level. Fighting an agent at 1/100 speed would be unplayable, but then again, thats the point - as you get closer to his ability you get more able to fight them; and you're not a normal human so you'd be moving about 1/10 agent speed, not 1/100. Scaling the effect up to 100m from the person in bullet time so that v -> v/(1+f(100-r)), where f=1 (1/100 speed at centre) for an agent, the bullet takes 10s to hit from 100m. Putting f=0.5 (1/10 speed at centre, more like a good normal player) a bullet still takes 5s to hit from 100m. Even a spray of bullets from the edge of the bubble would be dodgeable.
There are intrinsic problems with having multiple time scales in a multiplayer game, and those are a few of them. If they can pull it off, I will be very, very impressed.
Indeed. But the problems aren't intractable. The biggest difficulty is making this a game worth playing; the console matrix game was totally crap, remember.
Its not/that/ hard to do. Here's how: you can have two kinds of bullet time, 'slow' time, where other characters get slowed down, and 'fast' time, where your character is speeded up.
For 'slow' time, the effect is confined to a bubble, so you can have bullet time fights without affecting the whole world. As different characters are differently able to distort time, control speed and physics are changed by different amounts depending on your abilities; but all characters 'see' the bubble as if they were the fastest character (ie you see it the way the movie did).
To avoid discontinuities, intra-world communication (eg speech, phone) with someone in a different timeframe is disrupted, but inter-world communication is unaffected (eg a 'find me an exit' conversation). Constant rain and other effects can be used to constrain line-of-sight so there are no visual discontinuities.
'fast' time can be used for pre-planned actions, eg running along a route. Since you don't need real-time control, a fast character can then be moved at super-speed to their destination, and slow characters see an appropriate visual effect. Encountering a 'slow' bubble while moving along a 'fast' route drops you back into control.
I'm sure this leads to plenty of temporal anomalies, but unless you give they characters wristwatches they won't be annoyingly obvious.
From the pictures on the site, I imagine that they are just going to make a full piece of cement to your size specifications insted of you buying a whole bunch of smaller blocks and cementing them.
No, this picture clearly shows bricks. However generally cement isn't going to be a problem for this material. You can see that he's running fibres from one site of the concrete to the other. The light you see on one side appears in the position it fell on the other side, (look at the two shadows of the woman's right arm - the one cast outside the block is lower than the one cast through the block).
This being the case,you can create room for a channel of cement between blocks by bending the fibres, while having the blocks appear to be flush on the outside. ie:
yup its pretty disgusting. I havent actually done that since some all nighters 10 years ago (reheating filter coffee rather than waste energy making a whole new pot). Apparently you can get decent coffee this way though, using a cold brew process.
Anyway I don't need a command line for making coffee - its already online.
But you think a command line would make using a digicam easier? A microwave? A thermostat?
For 2 out of 3 of these, the answer is yes. I'd probably say yes to all 3 if I used a digicam more often.
Hardware user interfaces are patentable (do a USPTO search for microwave control panel) and thus there is a small fee for including one developed elsewhere. For low-end consumer devices, with small margins, this introduces a disincentive to using the best user interface. As a result, microwave UIs are generally confusing. I often see people in our office push buttons semi-randomly to get the thing to start, then open the door to stop it when enough seconds have passed - they can't even operate the timer.
Similarly, looking at the timed thermostat in my house, it uses little pushpins to determine on/off times, but its difficult to set the/current/ time, the power level doesn't relate to the temperature, and all the radiators have their own thermostat anyway. Its a mess.
Microwave >cook 30s >cook full power 30s >defrost 1m 30s >at 08:00 cook 30s #breakfast coffee? >in 35m cook 3m #cook veg when roast is ready
Thermostat >list sensors >alias bedroom sensor3 >at 07:00 bedroom 22c #before I wake >at 09:00 bedroom off #after I leave >at 19:00 tvroom 20c >house warmer 2c
It stands for Document Type Definition, and includes all the human-readable specification that describes what all the elements mean as well as the formal, machine-readable part of the specification
Hmm. You mean it has comments? Because a DTD is nothing more than a machine-readable specification, which often (but not always) comes with comments.
If you somehow came to the conclusion that xml schemas (in general) are not meant to be human readable, take a look at the compact syntax for RNG.
Norm Walsh (ie Mr DocBook) is already making progress replacing the DTD infrastructure of DocBook with RNG. And guess what, he uses an editor (nXML-mode for GNU Emacs) that supports RNG!
I guess he must wear a blindfold? Or maybe you should take a read of James Clark's paper on the design of Relax NG?
It's easy to ctrl-click with one hand - hold down ctrl with your left hand, and bang your head repeatedly on the mouse button.
As a fairly recent switcher too, I bought Warcraft III with the powerbook so I could get into the way of the one button mouse with some fun rather than frustrated attempts at work. Its pretty much second nature now. One thing you notice is that you don't need to right click much anyway - OS X doesn't emphasise context menus as much as Windows.
Clicking belongs to the mouse. Using the keyboard makes it all more confusing. Face it, a two button mouse will always be better than one button mouse.
Under windows I use a 5 button MS optical trackball, with a scrollwheel. I have them mapped to Escape, Meta, Alt, Control, and Shift[1]. Because none of those belong on a keyboard;)
-Baz
[1] In case you're wondering, the scrollwheel adjusts the strength on the coffee machine. I installed a ratchet so it only turns one way.
Re:Today only, free access courtesy of Slashdot
on
Orwellian Tech Support
·
· Score: 3, Funny
Which Tech Support Staffer Are You?
I think I am a "Santa"
You're sacked. I on the other hand am Rudolf the "who knows?" Reindeer.
Read this - a press release from the US army STRICOM dated Nov. 19th. 2003 - there was probably some other US coverage at that time. The article's a bit more informative than the beeb one actually, as it shows the size of the There contract ($3.5m - which I guess puts it as something between 6mo. and a year, depending on the team size - its interesting that There haven't even put a press release about it), and that the Army are funding this speculatively - there is no group that actually wants this for training yet.
The beeb is reporting it because they read the article on Homelan Fed last week. There's more coverage here
Oh fer crying out loud moderators, the parent is not offtopic. Whether it's funny or overrated is up to you.
Anyway, I'm left wondering at the lack of tabloid sensibilities in the journos covering this: "New star emerges from dust cocoon" (BBC) "Amateur Astronomer Discovers New Nebula" (/.)
Erm, great headlines guys. What the heck happened to "A Star Is Born"? "Twinkle twinkle little star"? "Every nebula has a silver lining?" or even "A long time ago and far, far away"?
If you look at what is proposed, it describes clients sending tokens like this:
GET http://www.webserver.com/webpage.html HTTP/1.1 Authorization: JabberTicket 54yudvjhssa76dta6sgdst78r4sadsfjdhs...
now apart from the nitpicking complaint that they should use example.com as the test domain (follow link to see why), its obvious that this needs client-side support. With browser rollouts being mindnumbingly s l o w, that means they are probably targeting web services, or non-browser clients, or must be building a browser extension?
Secondly, the spec for the client request for a ticket doesnt include any authentication info whatsoever. Ok, this means they must be doing that in 'some other protocol' (presumably Jabber + SASL). They could be a bit clearer... this part basically requires you to have a fairly complete XMPP implementation in order to get at the apparently simple ticket service.
Mark me down as unconvinced. Take a look at Shibboleth and OpenSAML to see what others are doing in this space - they are already doing single sign on, and it already works (OpenSAML does have the downside of being affected by a free-to-license RSA patent).
We have integrated sites into Athens (SSO for the UK EDU/GOV sectors), which is similar to Shibboleth in scope, and doesn't require browser changes.
Just to repeat what everyone else says first: its very important to design to standards. Its the only way to be sure that things will continue to work.
But a second point: many browser statistics are misleading. As well as it being easier for non-IE users to block webbugs and fake user-agent strings, sites that are designed for IE only suck in other browsers, and arent 'sticky' for non IE users. Non IE users will be consistently under-reported on many sites as a consequence.
So, if there are 10% of users using non-IE, and you are only getting half of them, should you care? Well, yes. Non-IE users are more likely to be technically savvy - they've got as far as changing their browser!.. and are thus more likely to be educated, employed and able to pay for stuff (especially small shiny expensive stuff). So you lost 50% of one of the best market sectors, while keeping 100% of the deadbeats. Nice move.
Knowing your market sector is more important than knowing what your browser shares have been. E.g. we work in on sites which are largely browsed by architects, and guess what - they largely use Macs. But we didn't get them until we spent time making the site Mac friendly.
Here's the SciMark 2 benchmark scores for various platforms. This isn't an entirely relevant comparison for development as its comparing one aspect of CPU+JVM, eg I/O is neglible, but it gives you some idea of whats going available.
A dual G5 placed 15th in the list of submitted scores: 1. 554.92 Sun 1.4.2 WinXP 5.1 Sun. 1.4.2; IBM; P4 3GHz [...] 15. 226.23 Apple 1.4.1_01 MacOSX 0.2.7 Apple Computer, Inc. 1.4.1_01; Apple G5; PowerPC 970 2x2Ghz
there's a lot of x86 ahead of PPC. I've got no axe to grind here, I have a powerbook myself and I find it a pretty decent working environment - partly I wanted one because I use wintel in work and wanted to get away from that when doing my own stuff.
RSS feeds are animal byproducts made according to the Revised Statutes of Saskatchewan (RSS). In the case of websites, mostly it refers to recycled bovine ordure, or "bull logs", often pronounced colloquially as "blogs".
Part of the problem here is that there isn't enough mindshare for the idea that the format you edit in isn't the format you share files in. MS are at least indirectly to blame here, their monopolistic status has given people the impression that MS Office formats are appropriate formats to share documents.
In the UK govt software market, people who make systems to support the Public Records Office standard - for archiving, ie sharing documents with our future selves - must support storing the original document, AND a 'rendered' copy that will still be legible in future.
Looking at them from this archival viewpoint, it becomes obvious that MS Office formats are inappropriate for rendered/interchange usage, as OLE containers can contain pretty much anything - you need every windows program on the planet to be sure you can read them. Not only that, they often preserve editing/versioning information which you don't want to share (see eg. the leaked origins of the UK WMD Dossier).
I don't expect things to change anytime soon, theres no money in it; and none of the available options are all that great (PDF also has plugins, PS isnt supported well in windows, XML can contain arbitrary meaningless binary data, etc). For now, the best you can do is to aim at xhtml+jpg as your interchange format, producing presentations using e.g. slideml.
Owen Densmore covered this in his O'Reilly blog last year - he checked/usr/dict/words against DNS, to see how many words werent taken. There were only 43 4-letter words left from the.com namespace, junk like "frib", and "odso".
There were a few thousand 5- and 6- letter words left, but again, all pretty uncommon words: "upwaft.com" or "bepity.com" anyone? Most 'real' words are claimed by someone, somewhere, and the only option for making a name that uses words people know is to make one up by sticking words together, or letters and words together.
But I've been to two Legolands, and I knew better than to even consider applying.
Like the other guy said, why not? When I was applying for my first graduate job, Biddy Baxter retired - if you're not from the UK or don't remember, she was the producer for Blue Peter and for years came up with all the crappy toys they'd show you how to build with yoghurt pots and sticky back plastic. Her job was advertised in the Guardian so of course I applied....never had a hope in hell of getting it but it makes a change from going for crappy office jobs.
Mod parent up, doing this with CDs would be painful and time consuming. I used to work at a facility managing firewalls worldwide for a multinational, our requirements for regular machine resets weren't all that dissimilar from a testbed (although on a larger scale) - and this is exactly how it was done.
Another alternative to consider, which especially considering your mixed environment, might be better for you, is virtualization - e.g. using VMWare as described here. This avoids the need to image the machine entirely.
I took a gander at the delegated registrars, and one of them is Russ Nelson (of Crynwr driver fame, and who has in the past been a regular poster to/.)
NY ALEXANDRIA-BAY.NY.US. nelson@crynwr.com
Since this is clearly an "individual geek" operation like your own, he might be able to help.
Programmers are nothing like actors - game characters are like actors. As in, I liked Mario's earlier, funnier work.
The superstar programmers are more like the auteurs of old, doing script, direction, cinematography and editing - but not acting - and even in gaming they are the exception not the rule.
Its possible that soon actors will be like actors - games already contain their mocap'd, scanned, and sampled avatars. How long will it be before having an actor in a game guarantees an opening weekend?
This would require the respondant to flag which part of the email he's responding to, no? And he'd have to do that multiple times to respond to multiple points, no?
No flags are required. The clues are there in human-readable form, and for the most part are machine-readable, given enough smarts in the parser. While generally responses may be entirely above the message they reply to, or entirely below, enough are 'piecewise below' to be useful, and its possible to identify the pieces by looking for how the text was quoted (as supercite does, with regexps) or by looking at the original message - since I'm talking about summarising whole threads, I always know what the original message said.
That's probably more effort than almost anyone will expend. In/. and other fora that don't let you automatically quote the original it is stupidly hard, yes, but the usage is common on any mailing list because email clients are much better. Mailing lists also have a low incidence of html mail, which would be a big problem for me discovering responses.
I've found that if you include multiple points in your message you'll get a response on only one. Not from me;)
I don't see a piece of software, particularly a Microsoft piece of software, making this easy enough for the masses any time soon. I agree with you entirely. However the idea has pure geek value, which makes it a tad more likely to get done.
I'm not sure if MS aren't talking about something different from what most of this discussion thinks they are. Rather than showing the thread of discussion of whole emails (which we're all used to in other clients) it might be they mean something more like this old discussion of what e-mail discussions should look like by Ka Ping-Yee..
In case you manage to/. that, the idea is that it shows the responses to pieces of your email - the kind where someone says "see my responses inline" and responds to each of your points piecemeal, then you do the same to their responses, and so on.
I've often thought it would be cool to write something to parse emails the KPY way, but the heuristics would have to be pretty damn clever to deal with supercite. Specifically what I wanted was something that combined KPY's ideas with text-autosummarization , and some 'author ranking' information to produce mailing list summaries from gmane which are like Kernel Traffic and Cousins, or the now-defunct Eclectic.
Oh well, I can always wait until MS put this in Outlook 2010;)
Slashdot Mirror
So, let me try to understand your vision: Players are running around shooting each other in real time, but someone is engaged in a bullet-time battle, they are shown as being very slow. However, this won't affect the rest of the world?
Anyone who can see into the bubble is part of the bubble, was what I meant. That's why I mentioned the rain; you have to constrain line-of-sight so the effect doesn't spread to the whole MMO world.
Also, what if you're in a team? You have five characters (Team A) who encounter 5 other characters (Team B). Two opposing characters, one from each team, go into bullet-time.
The players are all involved in the bubble. The players without bullet time enabled have slowed controls and move like they're stuck in mud. The characters with bullet time enabled move normally except that they appear less affected by gravity (9.8 ms-2 appears as 0.98ms-2 if one 'real' second is 0.1 'in-game' seconds)
Alternatively, while the two players are in bullet time, you take your four team members and spray as much machine-gun fire into the bubble as possible...
etc, etc - you've misunderstood. All players within sight of each other are affected by the bubble.
your control is stopped immediately when someone pops into bullet time, then you've just given controllable lag to upper-level characters
Yes, precisely (except that control is slowed, not stopped). What do you think bullet time is, but the ability for agents to avoid the lag of the real world? However, if you have any ability available, turning on your ability should be either automatic or instantaneous, not delayed as for normal motion.
Then you have to determine what the affected radius of bullet time exposure is, whether it works through walls and windows, etc.
You can have it purely line of sight, but that has obvious problems with 'infecting' the whole world with slowness, unless you prevent people forming a human chain of lines of sight to the original bubble.
My preferred alternative would be to make the effect depend on your radial distance from a character in bullet time. In this case, yes, bullets from outside would slow as they neared the location of characters in bullet time; but if the radius of effect is large enough - say linearly decreasing to 100m - you still have plenty of time to see bullets coming.
Putting some real figures on this - 500m/s muzzle velocity is realistic, 5m/s is dodgeable so 1/100 speed is agent-level. Fighting an agent at 1/100 speed would be unplayable, but then again, thats the point - as you get closer to his ability you get more able to fight them; and you're not a normal human so you'd be moving about 1/10 agent speed, not 1/100. Scaling the effect up to 100m from the person in bullet time so that v -> v/(1+f(100-r)), where f=1 (1/100 speed at centre) for an agent, the bullet takes 10s to hit from 100m. Putting f=0.5 (1/10 speed at centre, more like a good normal player) a bullet still takes 5s to hit from 100m. Even a spray of bullets from the edge of the bubble would be dodgeable.
There are intrinsic problems with having multiple time scales in a multiplayer game, and those are a few of them. If they can pull it off, I will be very, very impressed.
Indeed. But the problems aren't intractable. The biggest difficulty is making this a game worth playing; the console matrix game was totally crap, remember.
Its not /that/ hard to do. Here's how: you can have two kinds of bullet time, 'slow' time, where other characters get slowed down, and 'fast' time, where your character is speeded up.
For 'slow' time, the effect is confined to a bubble, so you can have bullet time fights without affecting the whole world. As different characters are differently able to distort time, control speed and physics are changed by different amounts depending on your abilities; but all characters 'see' the bubble as if they were the fastest character (ie you see it the way the movie did).
To avoid discontinuities, intra-world communication (eg speech, phone) with someone in a different timeframe is disrupted, but inter-world communication is unaffected (eg a 'find me an exit' conversation). Constant rain and other effects can be used to constrain line-of-sight so there are no visual discontinuities.
'fast' time can be used for pre-planned actions, eg running along a route. Since you don't need real-time control, a fast character can then be moved at super-speed to their destination, and slow characters see an appropriate visual effect. Encountering a 'slow' bubble while moving along a 'fast' route drops you back into control.
I'm sure this leads to plenty of temporal anomalies, but unless you give they characters wristwatches they won't be annoyingly obvious.
No, this picture clearly shows bricks. However generally cement isn't going to be a problem for this material. You can see that he's running fibres from one site of the concrete to the other. The light you see on one side appears in the position it fell on the other side, (look at the two shadows of the woman's right arm - the one cast outside the block is lower than the one cast through the block).
This being the case,you can create room for a channel of cement between blocks by bending the fibres, while having the blocks appear to be flush on the outside. ie:
yup its pretty disgusting. I havent actually done that since some all nighters 10 years ago (reheating filter coffee rather than waste energy making a whole new pot). Apparently you can get decent coffee this way though, using a cold brew process.
Anyway I don't need a command line for making coffee - its already online.
But you think a command line would make using a digicam easier? A microwave? A thermostat?
/current/ time, the power level doesn't relate to the temperature, and all the radiators have their own thermostat anyway. Its a mess.
For 2 out of 3 of these, the answer is yes. I'd probably say yes to all 3 if I used a digicam more often.
Hardware user interfaces are patentable (do a USPTO search for microwave control panel) and thus there is a small fee for including one developed elsewhere. For low-end consumer devices, with small margins, this introduces a disincentive to using the best user interface. As a result, microwave UIs are generally confusing. I often see people in our office push buttons semi-randomly to get the thing to start, then open the door to stop it when enough seconds have passed - they can't even operate the timer.
Similarly, looking at the timed thermostat in my house, it uses little pushpins to determine on/off times, but its difficult to set the
Microwave
>cook 30s
>cook full power 30s
>defrost 1m 30s
>at 08:00 cook 30s #breakfast coffee?
>in 35m cook 3m #cook veg when roast is ready
Thermostat
>list sensors
>alias bedroom sensor3
>at 07:00 bedroom 22c #before I wake
>at 09:00 bedroom off #after I leave
>at 19:00 tvroom 20c
>house warmer 2c
I think /you/ are completely confused.
It stands for Document Type Definition, and includes all the human-readable specification that describes what all the elements mean as well as the formal, machine-readable part of the specification
Hmm. You mean it has comments? Because a DTD is nothing more than a machine-readable specification, which often (but not always) comes with comments.
If you somehow came to the conclusion that xml schemas (in general) are not meant to be human readable, take a look at the compact syntax for RNG.
Norm Walsh (ie Mr DocBook) is already making progress replacing the DTD infrastructure of DocBook with RNG. And guess what, he uses an editor (nXML-mode for GNU Emacs) that supports RNG!
I guess he must wear a blindfold? Or maybe you should take a read of James Clark's paper on the design of Relax NG?
It's easy to ctrl-click with one hand - hold down ctrl with your left hand, and bang your head repeatedly on the mouse button.
;)
As a fairly recent switcher too, I bought Warcraft III with the powerbook so I could get into the way of the one button mouse with some fun rather than frustrated attempts at work. Its pretty much second nature now. One thing you notice is that you don't need to right click much anyway - OS X doesn't emphasise context menus as much as Windows.
Clicking belongs to the mouse. Using the keyboard makes it all more confusing.
Face it, a two button mouse will always be better than one button mouse.
Under windows I use a 5 button MS optical trackball, with a scrollwheel. I have them mapped to Escape, Meta, Alt, Control, and Shift[1]. Because none of those belong on a keyboard
-Baz
[1] In case you're wondering, the scrollwheel adjusts the strength on the coffee machine. I installed a ratchet so it only turns one way.
Which Tech Support Staffer Are You?
I think I am a "Santa"
You're sacked. I on the other hand am Rudolf the "who knows?" Reindeer.
Read this - a press release from the US army STRICOM dated Nov. 19th. 2003 - there was probably some other US coverage at that time. The article's a bit more informative than the beeb one actually, as it shows the size of the There contract ($3.5m - which I guess puts it as something between 6mo. and a year, depending on the team size - its interesting that There haven't even put a press release about it), and that the Army are funding this speculatively - there is no group that actually wants this for training yet.
The beeb is reporting it because they read the article on Homelan Fed last week. There's more coverage here
Oh fer crying out loud moderators, the parent is not offtopic. Whether it's funny or overrated is up to you.
Anyway, I'm left wondering at the lack of tabloid sensibilities in the journos covering this:
"New star emerges from dust cocoon" (BBC)
"Amateur Astronomer Discovers New Nebula" (/.)
Erm, great headlines guys. What the heck happened to "A Star Is Born"? "Twinkle twinkle little star"? "Every nebula has a silver lining?" or even "A long time ago and far, far away"?
-"Lunchtime O'Booze"
If you look at what is proposed, it describes clients sending tokens like this:
GET http://www.webserver.com/webpage.html HTTP/1.1
Authorization: JabberTicket 54yudvjhssa76dta6sgdst78r4sadsfjdhs...
now apart from the nitpicking complaint that they should use example.com as the test domain (follow link to see why), its obvious that this needs client-side support. With browser rollouts being mindnumbingly s l o w, that means they are probably targeting web services, or non-browser clients, or must be building a browser extension?
Secondly, the spec for the client request for a ticket doesnt include any authentication info whatsoever. Ok, this means they must be doing that in 'some other protocol' (presumably Jabber + SASL). They could be a bit clearer... this part basically requires you to have a fairly complete XMPP implementation in order to get at the apparently simple ticket service.
Mark me down as unconvinced. Take a look at Shibboleth and OpenSAML to see what others are doing in this space - they are already doing single sign on, and it already works (OpenSAML does have the downside of being affected by a free-to-license RSA patent).
We have integrated sites into Athens (SSO for the UK EDU/GOV sectors), which is similar to Shibboleth in scope, and doesn't require browser changes.
Just to repeat what everyone else says first: its very important to design to standards. Its the only way to be sure that things will continue to work.
But a second point: many browser statistics are misleading. As well as it being easier for non-IE users to block webbugs and fake user-agent strings, sites that are designed for IE only suck in other browsers, and arent 'sticky' for non IE users. Non IE users will be consistently under-reported on many sites as a consequence.
So, if there are 10% of users using non-IE, and you are only getting half of them, should you care? Well, yes. Non-IE users are more likely to be technically savvy - they've got as far as changing their browser!.. and are thus more likely to be educated, employed and able to pay for stuff (especially small shiny expensive stuff). So you lost 50% of one of the best market sectors, while keeping 100% of the deadbeats. Nice move.
Knowing your market sector is more important than knowing what your browser shares have been. E.g. we work in on sites which are largely browsed by architects, and guess what - they largely use Macs. But we didn't get them until we spent time making the site Mac friendly.
-Baz
Here's the SciMark 2 benchmark scores for various platforms. This isn't an entirely relevant comparison for development as its comparing one aspect of CPU+JVM, eg I/O is neglible, but it gives you some idea of whats going available.
A dual G5 placed 15th in the list of submitted scores:
1. 554.92 Sun 1.4.2 WinXP 5.1 Sun. 1.4.2; IBM; P4 3GHz
[...]
15. 226.23 Apple 1.4.1_01 MacOSX 0.2.7 Apple Computer, Inc. 1.4.1_01; Apple G5; PowerPC 970 2x2Ghz
there's a lot of x86 ahead of PPC. I've got no axe to grind here, I have a powerbook myself and I find it a pretty decent working environment - partly I wanted one because I use wintel in work and wanted to get away from that when doing my own stuff.
RSS feeds are animal byproducts made according to the Revised Statutes of Saskatchewan (RSS). In the case of websites, mostly it refers to recycled bovine ordure, or "bull logs", often pronounced colloquially as "blogs".
Hope that clears things up.
Part of the problem here is that there isn't enough mindshare for the idea that the format you edit in isn't the format you share files in. MS are at least indirectly to blame here, their monopolistic status has given people the impression that MS Office formats are appropriate formats to share documents.
In the UK govt software market, people who make systems to support the Public Records Office standard - for archiving, ie sharing documents with our future selves - must support storing the original document, AND a 'rendered' copy that will still be legible in future.
Looking at them from this archival viewpoint, it becomes obvious that MS Office formats are inappropriate for rendered/interchange usage, as OLE containers can contain pretty much anything - you need every windows program on the planet to be sure you can read them. Not only that, they often preserve editing/versioning information which you don't want to share (see eg. the leaked origins of the UK WMD Dossier).
I don't expect things to change anytime soon, theres no money in it; and none of the available options are all that great (PDF also has plugins, PS isnt supported well in windows, XML can contain arbitrary meaningless binary data, etc). For now, the best you can do is to aim at xhtml+jpg as your interchange format, producing presentations using e.g. slideml.
Owen Densmore covered this in his O'Reilly blog last year - he checked /usr/dict/words against DNS, to see how many words werent taken. There were only 43 4-letter words left from the .com namespace, junk like "frib", and "odso".
There were a few thousand 5- and 6- letter words left, but again, all pretty uncommon words: "upwaft.com" or "bepity.com" anyone? Most 'real' words are claimed by someone, somewhere, and the only option for making a name that uses words people know is to make one up by sticking words together, or letters and words together.
-Baz
But I've been to two Legolands, and I knew better than to even consider applying.
Like the other guy said, why not? When I was applying for my first graduate job, Biddy Baxter retired - if you're not from the UK or don't remember, she was the producer for Blue Peter and for years came up with all the crappy toys they'd show you how to build with yoghurt pots and sticky back plastic. Her job was advertised in the Guardian so of course I applied....never had a hope in hell of getting it but it makes a change from going for crappy office jobs.
Icecast is a great piece of software.. I use it to stream music to myself when I am away from my PC.
I prefer micecast myself.
Yawwwwwwwwn!!!
I'm only using windows reluctantly but this is ridiculous. You can do the exact same thing with nslookup, supplied with windows:
G:\>nslookup
> set q=txt
> aol.com
Server: XXXXXXXXXXXXXXX
Address: XXXXXXXXXXXXXXX
Non-authoritative answer:
aol.com text =
"v=spf1 ip4:152.163.225.0/24 ip4:20....
Mod parent up, doing this with CDs would be painful and time consuming. I used to work at a facility managing firewalls worldwide for a multinational, our requirements for regular machine resets weren't all that dissimilar from a testbed (although on a larger scale) - and this is exactly how it was done.
Another alternative to consider, which especially considering your mixed environment, might be better for you, is virtualization - e.g. using VMWare as described here. This avoids the need to image the machine entirely.
I took a gander at the delegated registrars, and one of them is Russ Nelson (of Crynwr driver fame, and who has in the past been a regular poster to /.)
NY ALEXANDRIA-BAY.NY.US. nelson@crynwr.com
Since this is clearly an "individual geek" operation like your own, he might be able to help.
Sorry Russ if this is unwelcome!
Really, you shouldn't use the word 'degree' with the work 'Kelvin' as in the case used in the Story. It's preferable to write simply 5789 Kelvin.
Since we're nitpicking... its kelvin, not Kelvin.
Programmers are nothing like actors - game characters are like actors. As in, I liked Mario's earlier, funnier work.
The superstar programmers are more like the auteurs of old, doing script, direction, cinematography and editing - but not acting - and even in gaming they are the exception not the rule.
Its possible that soon actors will be like actors - games already contain their mocap'd, scanned, and sampled avatars. How long will it be before having an actor in a game guarantees an opening weekend?
This would require the respondant to flag which part of the email he's responding to, no? And he'd have to do that multiple times to respond to multiple points, no?
/. and other fora that don't let you automatically quote the original it is stupidly hard, yes, but the usage is common on any mailing list because email clients are much better. Mailing lists also have a low incidence of html mail, which would be a big problem for me discovering responses.
;)
No flags are required. The clues are there in human-readable form, and for the most part are machine-readable, given enough smarts in the parser. While generally responses may be entirely above the message they reply to, or entirely below, enough are 'piecewise below' to be useful, and its possible to identify the pieces by looking for how the text was quoted (as supercite does, with regexps) or by looking at the original message - since I'm talking about summarising whole threads, I always know what the original message said.
That's probably more effort than almost anyone will expend.
In
I've found that if you include multiple points in your message you'll get a response on only one.
Not from me
I don't see a piece of software, particularly a Microsoft piece of software, making this easy enough for the masses any time soon.
I agree with you entirely. However the idea has pure geek value, which makes it a tad more likely to get done.
-Baz
I'm not sure if MS aren't talking about something different from what most of this discussion thinks they are. Rather than showing the thread of discussion of whole emails (which we're all used to in other clients) it might be they mean something more like this old discussion of what e-mail discussions should look like by Ka Ping-Yee..
/. that, the idea is that it shows the responses to pieces of your email - the kind where someone says "see my responses inline" and responds to each of your points piecemeal, then you do the same to their responses, and so on.
;)
In case you manage to
I've often thought it would be cool to write something to parse emails the KPY way, but the heuristics would have to be pretty damn clever to deal with supercite. Specifically what I wanted was something that combined KPY's ideas with text-autosummarization , and some 'author ranking' information to produce mailing list summaries from gmane which are like Kernel Traffic and Cousins, or the now-defunct Eclectic.
Oh well, I can always wait until MS put this in Outlook 2010