So uh, I think a switch to wxWidgets or QT is the LAST thing they're gonna do, seeing as they created it in the first place.
What's lacking is a better native-toolkit engine for Windows and OSX, one that just shunts to the local system, rather than trying to emulate the local look. (On X11, you use the GTK engine the rest of your desktop is using, i.e. clearlooks)
Exactly. The user has no obligations. What crazy planet are you from where TCP/IP, peering agreements and ISPs work differently? When you put up a website, you are putting a SERVER out on the public internet which means you are making a calculated risk with known costs. Before you even lay out that initial month's payment you better have a fucking plan to make money that goes beyond serving banner ads.
The burden is _entirely_ with you to figure out how to make money from giving away data. You can't make any hard assumptions about user client behavior, only aggregate assumptions. Right now your stats have to assume the impression rate is much lower than before AdBlock (and similar tools) were popular.
Just because you're a webmaster and you're doing all this work doesn't necessarily mean your entitled to payment unless you're a salaried employee of a larger organization. The majority of webmasters who can live off their jobs maintain websites which are overall cost centers in their businesses which are not web-oriented. These websites DON'T make money directly; they enable the business the website is attached to make more money than they would without it.
But if you're going to go your own and make money purely through a website, unless you're selling a product or selling access, it's INCREDIBLY difficult.
You must deploy measures to deter non-money making uses of your resources, or only serve content to authenticated or paying users. Or be so insanely popular that advertisers throw money at your feet to appear up by your title line (Penny Arcade, Digg, etc.)
Users don't have to earn jack shit. This is where you are completely wrong.
As a webmaster, you bear ALL the burden. I wish webmasters would wake up and smell the fucking coffee.
* YOU (the webmaster) signed the agreement with your hosting provider (A dollars per month) * YOU (the webmaster) signed the agreement with the ad network (B dollars per impression) * YOU (the webmaster) are responsible for bringing to bear content that attracts visitors (C hits per month) * YOU (the webmaster) are responsible for technical measures that ensure that users can't get content without the ads (such that attrition rate k -> 0). * YOU (the webmaster) are responsible to maximize your own operating profit -- such that C*B*(1-k) >= A.
No where in this equation is the user expected to do anything. You bear all the risk. You are in control of A, B, C and k.
Wake up and smell the coffee, you whiny assholes. If you can't deal with this, then you need to get a new job you lazy PHP hacks.
Your next major home OS upgrade will cost you a bit per installation.
Frankly I'd be more supportive of MacOS if it was "sold" like Solaris is sold; you get a license for all supported versions with your hardware. Or that you could buy "N" upgrades upfront with hardware purchase for a reasonable fee (like 50% off).
$80 for what amounts to service packs is irritating to me.
We hooked it up to the audio in the house so we can hear it on the patio and you can control it with the BlueTooth remote from some distance away. DLNA (UPnP, whatever) hooked up to another machine in the house with all our music on it... That's a pretty nice thrown-in type of use for a game console. That and being able to watch MPEG4 video and use that same remote to shuttle it around without needing to boot it into some kind of other operating system for that kinda functionality.
Also, we have a lot of PS2 and PS1 games and it makes them look nice on the LCD with the HDMI output. Haven't run into compatibility issues yet.
If the other people who bought it were like us, it's because 1) We wanted a blue-ray/DVD player for an HD set. 2) We had a lot of PS2/PS1 games 3) We were going to take advantage of the media convergence features.
Sony was nice enough not to rope you into using their bullshit formats on #3. Straight MPEG-4 AVC video or MP3 audio on USB HDs or memory devices, from the internet, etc. etc... it's good enough even if they purposefully don't support MPEG-$ ASP (so you can't easily watch your pirated XviDs of Sony media properties, sigh).
I guess the point is you get a lot out of the box. Supported, no fiddling around with anything, and enough features unrestricted to keep you happy.
It struck me that the developers at Nintendo were using it as a platform to figure out control-scheme/scenarios that can be used in other games. Players who had this game early in the console's lifetime would, as a result, also be more prepared for future games if they used "strange" control mechanics.
Makes a ton of sense to me (fun game too). It made me go get the DS version.
Offtopic: Playing WarioWare got me thinking; there were mini games in which your hand motion with respect to the sensor bar allowed for precise positioning of an object in space that tracked very accurately, and since your hand's motion is limited by the weight of the wiimote it's as if there is a heft and physical extent to the virtual cursor as rendered in game. This is a very intuitive feeling and I understand why people gushed about Elebits in as far as the gameplay. You could keep using that metaphor in tons of games that have a physical aspect to them, so I expect to see a lot of 3d-physics type of interaction and problem solving which thankfully is in vogue now in game development libraries, and thus at developers' fingertips.
Only use the DNS naming specification for the share when you connect, i.e. \\foobar.example.com\sharename... it will use CIFS and not use netbios at all.
The same would be required of BSD, Solaris or Windows NT, neither of which have hard realtime scheduling either. Only OSX might be able to be used in such a fashion without virtualizing, if the Mach microkernel was hardened w.r.t. its realtime scheduling behavior.
I created a fake address just for dealing with them. I expected it to get spam the day I signed up. It's not really a legit operation, so would you expect them to give a shit about your contact info, I mean come on, like what are you going to do, call customer service and complain? Ha...
It powers Blue-Ray Disc menus/interactive content (BD-J), and is used in DVB and CableCard compliant set top boxes for interactive program guides, VoD ordering systems, etc.
You think they'll bring those back? And you'd think NVidia'd have the good sense to allow the nvidia kernel module to supply an fb block device for dicking around with virtual consoles and non-X guis. Ah well.
Move over Java... the new sandbox environment of this decade is the Win32 API running inside a virtual machine on linux or OSX. What software _doesn't_ work in such a setup aside from 3d games?
If you pay for Virtual Iron, then they hook you up with a better userspace abstraction component and paravirtualized drivers for Windows XP/2003 which are infinitely better than the emulated PCNet cards and stuff (even in HVM mode). Clever bastards.
We know this. The IOMMU work they are doing is _very_ interesting and I think is going to do that full circle thing (reliable hardware-assisted partitioning even in your sub-10k rackmount servers). They're in on the ground floor with this stuff. As soon as the management tools catch up, we'll be taking all this stuff for granted, and the old mainframe guys will just shake their heads.
... he'd be standing on top of a table right now screaming about something, NASA shuttle in space or not. He was a pretty intense kind of guy who could get away with standing on tables, soap boxes, and other tall things.
It's only two tiles that are damaged, but how big are they in the first place? They're not done running simulations for the effect on re-entry, but that non-smooth edge between the two damaged tiles in the gouge would worry me no matter the outcome with that much more friction and eddying.
Filesystems still allocate files with blocks that are multiples of 1k (1024). Binary numbers of blocks comprise larger allocation and accounting units and limits and such and so those binary representations are useful to system administrators who need to where those cylinder group boundaries are going to be... blah blah blah. Or what size blocks to use and how many when doing binary copies for restoring your data and such.
OTH you should be able to choose the representation. SI accounting for files > 1000000 bytes, and an integral number of 'blocks' for small files, whatever that smallest allocation unit is.
The definition for PCIe-2.0 is that the minimum duration in the asyncronously clocked 10b/8b encoded signal is 0.4 nanoseconds between transitions. The inverse (2.5GHz) gets divided by 10 to describe the maximum theoretical bytes per second (which is practically unattainable) and it gets multiplied by a thousand a few times to give people a number they can quote on message boards. So that's Hz based.
Memory is based on 100MHz clock multiples... blah blah blah
Modems -> telecom -> always used 10^3 multiples for everything
You're wrong on this one... bucking the storage trend these MBs are "real" MBs. 1024^2
SATA is based on a 1.5GHz signalling base (10b/8b, etc. etc.)
HT is similar to PCIe... blah blah blah
Yeah, we know about that one. That marketing crap (notice they don't get away with that shit with flash memory or USB memory sticks). Worse worse was floppies. 1.44MB = 1440 * 1024 bytes. Ugh.
So, uh, what non-Hz/telecom or non-marketing based ones again?
The inode representing the file and the file data itself are located at two different places on the disk. Both get read and cached when the file is opened and read.
Op1) Read inode into memory (cached) --- "Open" file for reading (inode gets atime updated, cached inode is now "dirty") Op2) Read requested blocks of file into memory (cached)... Some stuff happens with the file, and other files, or maybe nothing... OpN) bdflush getting impatient or a filesystem sync prompts the dirty inode to be written back out to disk and the memory reclaimed
So for every file read, there are three distinct operations, one at location A on the disk, another at location B, then a third at A again.
This is with normal atime behavior. Note that the operation at "N" is not done in a necessarily slow way; it might get batched in a bunch of delay writes to that region of the disk. But if your system isn't doing a bunch of reading and writing all over the disk, then this a very noticable seek in the grand scheme of things, especially if it's part of a linear search through files (like via grep or something).
This relatime patch changes things so that the atime is noted in memory, but the inode is not actually considered a candidate for updating "dirty" unless the PREVIOUS value of atime was at least a day ago. If the file is opened for writing, then the atime gets updated anyway along with the mtime; that always happens as you need to update the inode anyway when the length or data blocks change.
In the email discussion linked in the article kernel devs were wondering the same thing... why doesn't mutt/bash just use inotify on the mailbox/maildir if available at config/build time? (Linux does have a event system in epoll and inotify -- and we almost got that unified under kevent, and it's threatening to happen again pulling in AIO too)
They were also concerned about breakage of other systems that are not open source (HSMs) which use atime for usage tracking. Noting that HSMs and atime-sensitive maintenance programs usually don't exhibit time resolution in policies more fine than a day, they came up with the relatime = update at least once a day compromise.
Slashdot Mirror
and take your stupid me-too comment with you.
As if your opinion mattered... I'd love to see this wonderful code that you are somehow punishing us by not releasing under the BSD or GPL licenses.
You're forgetting something.
GTK stands for (drum roll) The Gimp ToolKit.
So uh, I think a switch to wxWidgets or QT is the LAST thing they're gonna do, seeing as they created it in the first place.
What's lacking is a better native-toolkit engine for Windows and OSX, one that just shunts to the local system, rather than trying to emulate the local look. (On X11, you use the GTK engine the rest of your desktop is using, i.e. clearlooks)
http://docs.sun.com/app/docs/doc/817-0547/eyaks?l=en&a=view&q=samba
You need to make sure all the SUNWsmb* packages are installed, then enable/configure libnss_winbind, pam_winbind and kerberos to talk to your AD.
It's a little tricky to set up but pretty straightforward if you know all your AD details.
What would be nice is some management tools that automate it but we can dream, right?
Exactly. The user has no obligations. What crazy planet are you from where TCP/IP, peering agreements and ISPs work differently? When you put up a website, you are putting a SERVER out on the public internet which means you are making a calculated risk with known costs. Before you even lay out that initial month's payment you better have a fucking plan to make money that goes beyond serving banner ads.
The burden is _entirely_ with you to figure out how to make money from giving away data. You can't make any hard assumptions about user client behavior, only aggregate assumptions. Right now your stats have to assume the impression rate is much lower than before AdBlock (and similar tools) were popular.
Just because you're a webmaster and you're doing all this work doesn't necessarily mean your entitled to payment unless you're a salaried employee of a larger organization. The majority of webmasters who can live off their jobs maintain websites which are overall cost centers in their businesses which are not web-oriented. These websites DON'T make money directly; they enable the business the website is attached to make more money than they would without it.
But if you're going to go your own and make money purely through a website, unless you're selling a product or selling access, it's INCREDIBLY difficult.
You must deploy measures to deter non-money making uses of your resources, or only serve content to authenticated or paying users. Or be so insanely popular that advertisers throw money at your feet to appear up by your title line (Penny Arcade, Digg, etc.)
Users don't have to earn jack shit. This is where you are completely wrong.
As a webmaster, you bear ALL the burden. I wish webmasters would wake up and smell the fucking coffee.
* YOU (the webmaster) signed the agreement with your hosting provider (A dollars per month)
* YOU (the webmaster) signed the agreement with the ad network (B dollars per impression)
* YOU (the webmaster) are responsible for bringing to bear content that attracts visitors (C hits per month)
* YOU (the webmaster) are responsible for technical measures that ensure that users can't get content without the ads (such that attrition rate k -> 0).
* YOU (the webmaster) are responsible to maximize your own operating profit -- such that C*B*(1-k) >= A.
No where in this equation is the user expected to do anything. You bear all the risk. You are in control of A, B, C and k.
Wake up and smell the coffee, you whiny assholes. If you can't deal with this, then you need to get a new job you lazy PHP hacks.
Hell, people were talking about this pie-in-the-sky shit since the late 90s, remember SimDesk?
Your next major home OS upgrade will cost you a bit per installation.
Frankly I'd be more supportive of MacOS if it was "sold" like Solaris is sold; you get a license for all supported versions with your hardware. Or that you could buy "N" upgrades upfront with hardware purchase for a reasonable fee (like 50% off).
$80 for what amounts to service packs is irritating to me.
We hooked it up to the audio in the house so we can hear it on the patio and you can control it with the BlueTooth remote from some distance away. DLNA (UPnP, whatever) hooked up to another machine in the house with all our music on it... That's a pretty nice thrown-in type of use for a game console. That and being able to watch MPEG4 video and use that same remote to shuttle it around without needing to boot it into some kind of other operating system for that kinda functionality.
Also, we have a lot of PS2 and PS1 games and it makes them look nice on the LCD with the HDMI output. Haven't run into compatibility issues yet.
If the other people who bought it were like us, it's because 1) We wanted a blue-ray/DVD player for an HD set. 2) We had a lot of PS2/PS1 games 3) We were going to take advantage of the media convergence features.
Sony was nice enough not to rope you into using their bullshit formats on #3. Straight MPEG-4 AVC video or MP3 audio on USB HDs or memory devices, from the internet, etc. etc... it's good enough even if they purposefully don't support MPEG-$ ASP (so you can't easily watch your pirated XviDs of Sony media properties, sigh).
I guess the point is you get a lot out of the box. Supported, no fiddling around with anything, and enough features unrestricted to keep you happy.
Warioware -- I've played it and enjoyed it.
It struck me that the developers at Nintendo were using it as a platform to figure out control-scheme/scenarios that can be used in other games. Players who had this game early in the console's lifetime would, as a result, also be more prepared for future games if they used "strange" control mechanics.
Makes a ton of sense to me (fun game too). It made me go get the DS version.
Offtopic: Playing WarioWare got me thinking; there were mini games in which your hand motion with respect to the sensor bar allowed for precise positioning of an object in space that tracked very accurately, and since your hand's motion is limited by the weight of the wiimote it's as if there is a heft and physical extent to the virtual cursor as rendered in game. This is a very intuitive feeling and I understand why people gushed about Elebits in as far as the gameplay. You could keep using that metaphor in tons of games that have a physical aspect to them, so I expect to see a lot of 3d-physics type of interaction and problem solving which thankfully is in vogue now in game development libraries, and thus at developers' fingertips.
I can't wait to see what comes of that.
Only use the DNS naming specification for the share when you connect, i.e. \\foobar.example.com\sharename ... it will use CIFS and not use netbios at all.
The same would be required of BSD, Solaris or Windows NT, neither of which have hard realtime scheduling either. Only OSX might be able to be used in such a fashion without virtualizing, if the Mach microkernel was hardened w.r.t. its realtime scheduling behavior.
I created a fake address just for dealing with them. I expected it to get spam the day I signed up. It's not really a legit operation, so would you expect them to give a shit about your contact info, I mean come on, like what are you going to do, call customer service and complain? Ha...
It powers Blue-Ray Disc menus/interactive content (BD-J), and is used in DVB and CableCard compliant set top boxes for interactive program guides, VoD ordering systems, etc.
You think they'll bring those back? And you'd think NVidia'd have the good sense to allow the nvidia kernel module to supply an fb block device for dicking around with virtual consoles and non-X guis. Ah well.
Move over Java... the new sandbox environment of this decade is the Win32 API running inside a virtual machine on linux or OSX. What software _doesn't_ work in such a setup aside from 3d games?
If you pay for Virtual Iron, then they hook you up with a better userspace abstraction component and paravirtualized drivers for Windows XP/2003 which are infinitely better than the emulated PCNet cards and stuff (even in HVM mode). Clever bastards.
We know this. The IOMMU work they are doing is _very_ interesting and I think is going to do that full circle thing (reliable hardware-assisted partitioning even in your sub-10k rackmount servers). They're in on the ground floor with this stuff. As soon as the management tools catch up, we'll be taking all this stuff for granted, and the old mainframe guys will just shake their heads.
... he'd be standing on top of a table right now screaming about something, NASA shuttle in space or not. He was a pretty intense kind of guy who could get away with standing on tables, soap boxes, and other tall things.
It's only two tiles that are damaged, but how big are they in the first place?
They're not done running simulations for the effect on re-entry, but that non-smooth edge between the two damaged tiles in the gouge would worry me no matter the outcome with that much more friction and eddying.
Filesystems still allocate files with blocks that are multiples of 1k (1024). Binary numbers of blocks comprise larger allocation and accounting units and limits and such and so those binary representations are useful to system administrators who need to where those cylinder group boundaries are going to be... blah blah blah. Or what size blocks to use and how many when doing binary copies for restoring your data and such.
OTH you should be able to choose the representation. SI accounting for files > 1000000 bytes, and an integral number of 'blocks' for small files, whatever that smallest allocation unit is.
Remember the uberlarge "32-megabit" SNES cartridges? 32 Mb = 4MB. A couple of floppies, which doesn't seem so impressive...
So, uh, what non-Hz/telecom or non-marketing based ones again?
The inode representing the file and the file data itself are located at two different places on the disk. Both get read and cached when the file is opened and read.
... Some stuff happens with the file, and other files, or maybe nothing...
Op1) Read inode into memory (cached)
--- "Open" file for reading (inode gets atime updated, cached inode is now "dirty")
Op2) Read requested blocks of file into memory (cached)
OpN) bdflush getting impatient or a filesystem sync prompts the dirty inode to be written back out to disk and the memory reclaimed
So for every file read, there are three distinct operations, one at location A on the disk, another at location B, then a third at A again.
This is with normal atime behavior. Note that the operation at "N" is not done in a necessarily slow way; it might get batched in a bunch of delay writes to that region of the disk. But if your system isn't doing a bunch of reading and writing all over the disk, then this a very noticable seek in the grand scheme of things, especially if it's part of a linear search through files (like via grep or something).
This relatime patch changes things so that the atime is noted in memory, but the inode is not actually considered a candidate for updating "dirty" unless the PREVIOUS value of atime was at least a day ago.
If the file is opened for writing, then the atime gets updated anyway along with the mtime; that always happens as you need to update the inode anyway when the length or data blocks change.
In the email discussion linked in the article kernel devs were wondering the same thing... why doesn't mutt/bash just use inotify on the mailbox/maildir if available at config/build time? (Linux does have a event system in epoll and inotify -- and we almost got that unified under kevent, and it's threatening to happen again pulling in AIO too)
They were also concerned about breakage of other systems that are not open source (HSMs) which use atime for usage tracking. Noting that HSMs and atime-sensitive maintenance programs usually don't exhibit time resolution in policies more fine than a day, they came up with the relatime = update at least once a day compromise.