I like the fact that Wikipedia is always current, so I don't know why I'd ever want this on DVD. For example, they had a great article on the pope the other day which was current right up through his death. Since I can just look this up online, why would I want stale information stored on my computer? I have a set of World Books on the shelf, and we keep them around for when you want to do research when "otherwise occupied" (i.e. sitting on the can). Of course now that I have a Zaurus with wireless networking then I don't need the hardcopy any more since I can surf from any room in the house.
I know that this is a little off-topic, but why do all of the 120 volt conversion when you can run directly off of low power? The display and keyboard sizes may pose a problem for you, but there has to be a solution somewhere. If I was going to a remote location then the last thing that I'd want to be thinking about is dragging along a heavy transformer/UPS and car battery.
Yeah, but the problem here is they serve both as a registrar and keeper of the registry. The only way to get rid of this problem is to split the two functions and prohibit one single company from doing both job functions. Kind of like the U.S. Mint - if you don't know how to make the paper *AND* the ink then you can't print your own money (unless you own a laser printer:-)
I agree with Swamii here. Just because people have strong religious views doesn't mean that they reject all science. I'm a pretty conservative guy and have lots of conservative friends, and I don't know any of them that would argue that there is life around volcanic vents in the deep ocean. Just because these guys can't make an interesting, profitable film doesn't mean that you should blame it on people who believe in the Bible.
I'm not being pedantic. As the "plantiff" in this case, you have the burden of proof that registries are better than storing files in/etc. So far your arguments have been:
Registries provide tools for parsing values from files
Registries provide centralized storage of data
Registries are easier to manage
That the Windows registry is sufficiently different from other registries to make any comparison invalid
I've already conceeded point #1 to you. I think we came to the agreement that point #2 isn't really unique to registries and therefore doesn't apply. The reason that I'm pressing on point #3 is that if you want me to take on a new tool then it had better solve the problem, if one exists. You haven't shown that there's a problem to begin with, or that cramming all config data into a registry fixes the "mess" of varying key structures within a node. In order for the registry to be stable and reliable for *every* application on your machine, I have to assume that I should never touch the internal data structures, which means that the Windows registry (#4) really isn't any different than any other registry on the planet.
If we agree on #1, why not just write the equivalent of GetPrivateProfileString() and be done with it? It seems like a much more balanced solution.
The fact that GConf uses XML is important not because the people should be editing its files by hand, but because it allows for simple recovery if something goes wrong.
Really? You just said that I should *never* be accessing the registry by any means other than the APIs or tools that the registry provides. So what does XML have to do with anything? Isn't the storage internal to the registry?
My mom isn't going to use gconf-editor. Nor is she going to use anything else. The end user could really care less how config data is stored, with the exception of wanting it backed up and easily restorable in the event of the failure. I don't get how a registry helps that. I can use tar to restore/etc as fast as any tool that you have can load the registry.
You didn't answer my comment about how one should use the registry. The problem isn't binary/non-binary, it's "am I going to use the registry or not". Using a registry for the sole reason of being able to easily parse data is not sufficient to make me want to switch. In addition, your implication that we bypass the registry in certain circumstances is very problematic. If I write data directly to files in the registry and then someone decides to change the way that the registry works then I've just introduced a new problem which my sysadmin isn't going to be very happy about. Of course you could always make the registry backward compatible on the storage side, but this seems like a waste of dev and test resources and leads to code bloat and performance loss.
For what it's worth, most admins would rather I write a sed script to update my.conf file than hand them a convoluted list of things that need to be updated in the registry. Across a large server farm, it's virtually impossible to make changes by hand across hundreds or thousands of servers.
Does anyone else have a problem that it's Friday night and we're arguing about the merits of registries?:-)
Why don't Windows registry complaints apply to GConf? If I'm using the registry "properly" then I should never access it outside of APIs and tools that are provided for management of the registry. The fact that GConf stores data in individual files that can be accessed from the filesystem is irrelevant. Either you chose to use the registry or you don't.
The main point of my first post is that centralizing all your config data into one tool doesn't add any value. Certainly centralizing it into one place makes backing stuff up easier, but you don't need a registry for that.
I know that some people love the idea of having a registry, but I personally think that it's a really bad idea. I've had plenty of experience with the Windows registry and I really hate it. What could be easier when configuring a machine than to copy a file from one place to another or just echo something into a config file? Yeah, regedit has some tools for loading.reg files into the registry, but why bother? And do you really think that the Windows registry is anything less of a mess than/etc? There's lots of undocumented crap in a Windows registry.
Supposedly registries are supposed to make administration easier, but in my experience they do exactly the opposite. Think about how many times you see instructions on Windows that say "before you make any changes, back up the registry". Having a registry makes sense when all your data all looks the same, but this isn't the case for config data. Every config file is different for a reason.
Of the many reasons that I switched from Windows to Linux, the ease of configuration was high on the list. I love RPMs because the config data just falls right into place without me having to touch anything or worry about messing up some other program's config data. Adding another layer to config data that doesn't add any value would just screw things up. I'd go so far as to argue that Windows would be a lot easier to deal with if they brought back.ini files.
I started on Mandrake at 8.0 after trying out FreeBSD, switched to Red Hat for a while because that's what data centers were running, and then went back to Mandrake when Red Hat killed support for 9.0. Mandrake is usually pretty stable and has lots of current versions of stuff, and I love the patch support for it. I signed up for the $120/yr MandrakeClub to help keep them in business, and I can download the newest PowerPack releases with BitTorrent and burn them to DVD before they're released to the public. They also have MandrakeMove which is fantastic for machine recovery if you do something stupid.
My only complaint is that Mandrake comes from France. It's not that I hate the French people (I love their fries;-), but rather that the language and time barriers sometimes make it difficult to interact with them.
Try Mandrake. Always current, great patch support, and urpmi is a dream. We're running it on production servers with virtually no problems. You'll want to shake out new versions before deploying them since the occasional issue crops up, but usually the next patch release fixes everything.
I'm curious to know how any organization can move forward when its leader can be replaced every year. A one year term gives you just enough time to get a good night's sleep between campaigns.
Back in the "good old days" when Netscape was the #1 browser, there was a term called "mind share" that BillG and others used to refer to the acceptance level that a company had within an industry. What has been killing Microsoft since the late 1990's is that it believes that its key market is corporations, but as we've all seen with Red Hat's near-abandonment of their user base is that corporations aren't the people that you really need to keep happy. You always have to please the end user, and the whole point of the blog is that the Web allows developers to interact directly with the end user without having to deal with internal bureaucracies and IT departments.
In some ways this the same evolution that we saw when moving from mainframes to PCs. When PCs weren't on the IT radar then you could do virtually anything you wanted, including upgrade the OS when you thought that it was worthwhile to do so. The problem that IT departments realized was that they were being squeezed out by the PC crowd so they started instituting controls which prevented the rapid upgrade cycle. This was good for the paid professionals that wanted to keep their jobs, but bad for the end user who just wanted a fix for the problem.
Microsoft's problem isn't so much that it can't ship software. Microsoft's problem is that it has the wrong target audience and an outdated delivery model. Microsoft was right to be nervous of the concept of the browser becoming the OS, and while it hasn't happened yet it appears that we're currently headed in that direction. I work for an ASP because I think that it's where the future is.
Let's face it, end users don't care about technology. They just want stuff that works. You can either stuff everything you'd ever want into a bloated box like Microsoft is doing today, or you can have a super-scalable solution where my web browser connects to servers that serve up specialized solutions. I may not get everything that I need, but at lot of small ASPs are going to be more responsive to their customers than large monolithic corporations like Microsoft. And they don't necessarily have to have "critical mass" to do so.
Actually, if you think about it, bloggers are going to burn books. Not literally, but figuratively. I think that this guy's concern about Google isn't that people are getting bad search results. His problem with Google is that I can now bypass my local library. Want to do some research? Do it online. Need access to an encyclopedia? Wikipedia will probably get you most of what you need to know. Libraries should embrace technology and there a lot that have. But at the end of the day the Internet could do to the library what it's done to the Post Office - stripped most of the revenue-generating material from their hands and left them with the leftovers that few are willing to pay for. (When was the last time you sent your grandmother a first-class piece of postal mail.) Let's face it, the only time I go to the library now is when I need some good reading material for the bathroom and I'm waiting on my next delivery of LinuxJournal or a book from Amazon.com.
I had the same problem. Thankfully, I had a solution. I had switched from Mandrake to RedHat about a year before then end of RHL9 on the theory that every data center in the world supported RedHat, but the switch back to Mandrake was painless. This is because Mandrake was originally based on RedHat, and virtually any RPM generated for RedHat runs on Mandrake. I get all the support that I want for free, although I subscribe to Mandrake Club for $120/year to help keep Mandrake in business. The distro is pretty stable and keeps current on most packages.
I'm not sure what this announcement means for us other than we might reconsider RedHat at some future point if they can step up to the plate. There's a lot for them to overcome though. Not only did we feel the pain of having them cut off support, but when we tried to complain via email we found that the first three email addresses we tried bounced back as invalid. If they can't keep valid email addresses on their web site, how much confidence can we have in these guys?
This is the same kind of argument that people use to justify overpriced hardware. I've seen roomfulls of expensive Compaq hardware because it had all kinds of wonderful features and was supposedly the "best". The problem is that you could engineer a fully redundant solution for less than 1/2 the hardware cost, and maintenance would be way fewer $ too. (Maybe you can ask Carly about this.) We run Mandrake for mission critical stuff and have very few problems. We occasionally get bitten by new bits, but then we just don't push the new release until it's full baked.
I stopped running RedHat when they dumped the old pricing model. I didn't mind paying them $, but I'm not a large corporation with money to burn either. Mandrake offers the same patch support that RedHat does and I've been running the 2.6 kernel for a very long time now.
So I guess that you wouldn't consider that the time that you spent asking your customers questions as to what they needed as "training time". After all, shouldn't you as a high-priced consultant know exactly what they need without you needing to ask? Or what about the countless hours that people spent writing HOWTOs, or the folks who go out of their way writing up those HTML pages that you Googled that told you everything that you needed to fix a problem, or those who write books? What if all these folks didn't want to part with their "hard earned knowledge"? With the exception of the books, how much of the information that you currently have stored in your head did you buy (including your own time spent playing with stuff) versus the amount that was freely given to you? When you do the calculation, make sure that you value the freely given information based on the real effort of the person who gave it to you, not how long it took you to read their article. I certainly hope you're not one of those guys who's never sent at least a "thank you" email to someone that wrote a HOWTO that saved you a bunch of time. I guess from your comments you probably click on every PayPal link on every page you come across because you so highly value everyone else's time as much as your own. (I'm not saying that I do, and quite frankly, I should do more of it.)
I don't disagree with you that there are leaches who will steal your ideas and resell them as their own, but I don't think that this poor schmuck was one of them. He's just a guy trying to support our cause and you're pissing in his coffee cup. You can't tell someone the RTFM when there's no FM to read.
I guess the difference between you and me is that I don't think of my knowledge as my own. I've worked my butt off for the last 28 years working on systems and at least 90% of what I know came from the hard work of other people. Most anyone who went to college owes their professors a debt of thanks given that a CS college professor's salary is roughly 50% of what they could make on the open market. My family gets a lot of credit as does yours, because they gave me the freedom to pursue my career and learn all this cool stuff. I'm sure that they'd be disappointed if I turned around and told them that they didn't get any credit for who and what I am today.
Man, I'm glad my doctor doesn't think like you. When I go to the doc, I tell him "Doc, my chest hurts". Now if I have a lung infection, would it be appropriate for my doctor to then tell me that I'm an idiot because I don't know the difference between my chest and lungs, and send me away with a harsh comment and a kick in the ass? What if I complained about a numb arm but I was really having a heart attack?
I get tired of reading crap like this from folks who "know better" than everyone else. I highly doubt that you were born with the knowledge between NTFS and Samba, which means that you possess your knowledge only because someone else was kind enough to pass along their understanding to you. So why do you repay other's kindness to you by calling someone else "stupid"? Is stroking your own ego more important than helping someone else who wants to learn something new?
"The other thing that occurred to me, is why do they feel like they have to upgrade?"
Hmmm, two things come to mind. #1 - For years, software vendors have been telling us that we need to update so that we get the latest and greatest features and for the most part we have bought into this theory whether it applies to us or not. #2 - Vulnerabilities in systems force them to be upgraded after their useful life expires. The definition of "useful life" is determined by the person who holds the code, unlike something like a car where the driver decides when to send it to the scrapyard.
Back in the good old days before C/C++ was popular, FORTRAN was a great programming language for doing anything that had to do with math or bit manipulation. The alternatives were things like COBOL or Pascal. What I loved about FORTRAN was that you could do cool things like treat strings as ints or do negative subscripting to rewrite the OS. I had a graphics class that I wrote some FORTRAN code in because I could break down a matrix into a single row just by passing in a pointer. I loved the fact that everything was pass by reference, which is why it took me so long to switch to C. If you had an optimizing compiler you could do stuff like make 1 == 2 because the constant value 1 was mapped to the same memory location for all instances and if you passed the constant into a function then you change the constant value inside the function.
I prided myself in college that I could write FORTRAN in any language. I had a prof that couldn't figure out why I was doing bit manipulation in COBOL. (Yes, this can be done in COBOL through multiplication and division, but it's really ugly.)
We use PHP for admin tasks all over the place, but the most useful place that we use it is in replicating our database. We have our master and slave server in place, but we didn't like MySQL's solution of replication because it assumes that the machines are visible to each other in a secure network enviroment. We replaced this with two PHP scripts that run in a cron job. On the master server we run "mysqladmin flush-logs" to close the current binlog and start a new one, and then the PHP script gzips and gpgs the binlog and puts it somewhere where we can suck it across the Internet. The slave machine pulls down the file, un-gpgs and ungzips it, and loads the file. Since we run this every 5 minutes, we never worry about losing a lot of work in our database.
For as much as I'd like to see Dino win, I gotta say that I'm really glad that we don't have Ron Sims running our state. This guy is about as liberal as they get, so having Chris Gregoire run the state doesn't seem all that bad. I agree with the previous post that the politicians don't really care as long as they win, which is why there are so many citizen initiatives on every ballot that we have here. The voters have already pushed down taxes several times and when the liberal judges of our fair state overturned the ballot initiatives then the legislature saw the light and passed their own version of the same laws. I guess what I'm saying is that I'm not too worried about it right now.
My company runs an ASP and as such we can send some pretty big pages (up to 500KB) back to the user. Gzip-compressed these pages are pretty small (50KB) so there's not a lot of time spent pulling the data across the wire (especially on broadband), so time to render on the client side is super important. We've seen that it can take several seconds for our pages to render on Mozilla, but in Firefox (as well as IE and Opera) they render in subsecond times. We also have a lot of javascript on the page, so I can't be sure whether its the decompression, table rendering, or javascript that's taking a long time in Mozilla and fixed in Firefox, but the difference can be clearly seen. Given that Firefox is the only real browser available for Linux and Mac right now, it's super important for us and our clients.
(Before I get flamed by the Konqueror/Safari crowd, let me say that there are lots of DHTML rendering problems on these browsers when it comes to repainting backgrounds and tables on the fly. Since we test in IE and Mozilla/Firefox and Opera and it all works great, I guess the problems must be with those browsers. Our Mac customers don't mind installing Firefox, so why should we bother trying to find workarounds?)
Having spent more than my fair share of time doing bug triage, I know how hard it can be to prioritize which bugs get fixed and in what order, hence my comment about making the hard choices. I don't blame anyone for the number of bugs, nor do I blame them for not cleaning everything up right away. My criticism is that new feature development takes priority over bug cleanup (thereby creating even more bugs), and for poor communication on what's happening and why.
Slashdot Mirror
I like the fact that Wikipedia is always current, so I don't know why I'd ever want this on DVD. For example, they had a great article on the pope the other day which was current right up through his death. Since I can just look this up online, why would I want stale information stored on my computer? I have a set of World Books on the shelf, and we keep them around for when you want to do research when "otherwise occupied" (i.e. sitting on the can). Of course now that I have a Zaurus with wireless networking then I don't need the hardcopy any more since I can surf from any room in the house.
What's the old joke? - "Never underestimate the bandwidth of a station wagon carrying magtapes".
I know that this is a little off-topic, but why do all of the 120 volt conversion when you can run directly off of low power? The display and keyboard sizes may pose a problem for you, but there has to be a solution somewhere. If I was going to a remote location then the last thing that I'd want to be thinking about is dragging along a heavy transformer/UPS and car battery.
Yeah, but the problem here is they serve both as a registrar and keeper of the registry. The only way to get rid of this problem is to split the two functions and prohibit one single company from doing both job functions. Kind of like the U.S. Mint - if you don't know how to make the paper *AND* the ink then you can't print your own money (unless you own a laser printer :-)
I agree with Swamii here. Just because people have strong religious views doesn't mean that they reject all science. I'm a pretty conservative guy and have lots of conservative friends, and I don't know any of them that would argue that there is life around volcanic vents in the deep ocean. Just because these guys can't make an interesting, profitable film doesn't mean that you should blame it on people who believe in the Bible.
- Registries provide tools for parsing values from files
- Registries provide centralized storage of data
- Registries are easier to manage
- That the Windows registry is sufficiently different from other registries to make any comparison invalid
I've already conceeded point #1 to you. I think we came to the agreement that point #2 isn't really unique to registries and therefore doesn't apply. The reason that I'm pressing on point #3 is that if you want me to take on a new tool then it had better solve the problem, if one exists. You haven't shown that there's a problem to begin with, or that cramming all config data into a registry fixes the "mess" of varying key structures within a node. In order for the registry to be stable and reliable for *every* application on your machine, I have to assume that I should never touch the internal data structures, which means that the Windows registry (#4) really isn't any different than any other registry on the planet.If we agree on #1, why not just write the equivalent of GetPrivateProfileString() and be done with it? It seems like a much more balanced solution.
Really? You just said that I should *never* be accessing the registry by any means other than the APIs or tools that the registry provides. So what does XML have to do with anything? Isn't the storage internal to the registry?
You didn't answer my comment about how one should use the registry. The problem isn't binary/non-binary, it's "am I going to use the registry or not". Using a registry for the sole reason of being able to easily parse data is not sufficient to make me want to switch. In addition, your implication that we bypass the registry in certain circumstances is very problematic. If I write data directly to files in the registry and then someone decides to change the way that the registry works then I've just introduced a new problem which my sysadmin isn't going to be very happy about. Of course you could always make the registry backward compatible on the storage side, but this seems like a waste of dev and test resources and leads to code bloat and performance loss.
For what it's worth, most admins would rather I write a sed script to update my .conf file than hand them a convoluted list of things that need to be updated in the registry. Across a large server farm, it's virtually impossible to make changes by hand across hundreds or thousands of servers.
Does anyone else have a problem that it's Friday night and we're arguing about the merits of registries? :-)
The main point of my first post is that centralizing all your config data into one tool doesn't add any value. Certainly centralizing it into one place makes backing stuff up easier, but you don't need a registry for that.
Supposedly registries are supposed to make administration easier, but in my experience they do exactly the opposite. Think about how many times you see instructions on Windows that say "before you make any changes, back up the registry". Having a registry makes sense when all your data all looks the same, but this isn't the case for config data. Every config file is different for a reason.
Of the many reasons that I switched from Windows to Linux, the ease of configuration was high on the list. I love RPMs because the config data just falls right into place without me having to touch anything or worry about messing up some other program's config data. Adding another layer to config data that doesn't add any value would just screw things up. I'd go so far as to argue that Windows would be a lot easier to deal with if they brought back .ini files.
My only complaint is that Mandrake comes from France. It's not that I hate the French people (I love their fries ;-), but rather that the language and time barriers sometimes make it difficult to interact with them.
Try Mandrake. Always current, great patch support, and urpmi is a dream. We're running it on production servers with virtually no problems. You'll want to shake out new versions before deploying them since the occasional issue crops up, but usually the next patch release fixes everything.
I'm curious to know how any organization can move forward when its leader can be replaced every year. A one year term gives you just enough time to get a good night's sleep between campaigns.
In some ways this the same evolution that we saw when moving from mainframes to PCs. When PCs weren't on the IT radar then you could do virtually anything you wanted, including upgrade the OS when you thought that it was worthwhile to do so. The problem that IT departments realized was that they were being squeezed out by the PC crowd so they started instituting controls which prevented the rapid upgrade cycle. This was good for the paid professionals that wanted to keep their jobs, but bad for the end user who just wanted a fix for the problem.
Microsoft's problem isn't so much that it can't ship software. Microsoft's problem is that it has the wrong target audience and an outdated delivery model. Microsoft was right to be nervous of the concept of the browser becoming the OS, and while it hasn't happened yet it appears that we're currently headed in that direction. I work for an ASP because I think that it's where the future is.
Let's face it, end users don't care about technology. They just want stuff that works. You can either stuff everything you'd ever want into a bloated box like Microsoft is doing today, or you can have a super-scalable solution where my web browser connects to servers that serve up specialized solutions. I may not get everything that I need, but at lot of small ASPs are going to be more responsive to their customers than large monolithic corporations like Microsoft. And they don't necessarily have to have "critical mass" to do so.
Actually, if you think about it, bloggers are going to burn books. Not literally, but figuratively. I think that this guy's concern about Google isn't that people are getting bad search results. His problem with Google is that I can now bypass my local library. Want to do some research? Do it online. Need access to an encyclopedia? Wikipedia will probably get you most of what you need to know. Libraries should embrace technology and there a lot that have. But at the end of the day the Internet could do to the library what it's done to the Post Office - stripped most of the revenue-generating material from their hands and left them with the leftovers that few are willing to pay for. (When was the last time you sent your grandmother a first-class piece of postal mail.) Let's face it, the only time I go to the library now is when I need some good reading material for the bathroom and I'm waiting on my next delivery of LinuxJournal or a book from Amazon.com.
I'm not sure what this announcement means for us other than we might reconsider RedHat at some future point if they can step up to the plate. There's a lot for them to overcome though. Not only did we feel the pain of having them cut off support, but when we tried to complain via email we found that the first three email addresses we tried bounced back as invalid. If they can't keep valid email addresses on their web site, how much confidence can we have in these guys?
I stopped running RedHat when they dumped the old pricing model. I didn't mind paying them $, but I'm not a large corporation with money to burn either. Mandrake offers the same patch support that RedHat does and I've been running the 2.6 kernel for a very long time now.
If RedHat is so great, why this versus this?
I don't disagree with you that there are leaches who will steal your ideas and resell them as their own, but I don't think that this poor schmuck was one of them. He's just a guy trying to support our cause and you're pissing in his coffee cup. You can't tell someone the RTFM when there's no FM to read.
I guess the difference between you and me is that I don't think of my knowledge as my own. I've worked my butt off for the last 28 years working on systems and at least 90% of what I know came from the hard work of other people. Most anyone who went to college owes their professors a debt of thanks given that a CS college professor's salary is roughly 50% of what they could make on the open market. My family gets a lot of credit as does yours, because they gave me the freedom to pursue my career and learn all this cool stuff. I'm sure that they'd be disappointed if I turned around and told them that they didn't get any credit for who and what I am today.
I get tired of reading crap like this from folks who "know better" than everyone else. I highly doubt that you were born with the knowledge between NTFS and Samba, which means that you possess your knowledge only because someone else was kind enough to pass along their understanding to you. So why do you repay other's kindness to you by calling someone else "stupid"? Is stroking your own ego more important than helping someone else who wants to learn something new?
Hmmm, two things come to mind. #1 - For years, software vendors have been telling us that we need to update so that we get the latest and greatest features and for the most part we have bought into this theory whether it applies to us or not. #2 - Vulnerabilities in systems force them to be upgraded after their useful life expires. The definition of "useful life" is determined by the person who holds the code, unlike something like a car where the driver decides when to send it to the scrapyard.
I prided myself in college that I could write FORTRAN in any language. I had a prof that couldn't figure out why I was doing bit manipulation in COBOL. (Yes, this can be done in COBOL through multiplication and division, but it's really ugly.)
We use PHP for admin tasks all over the place, but the most useful place that we use it is in replicating our database. We have our master and slave server in place, but we didn't like MySQL's solution of replication because it assumes that the machines are visible to each other in a secure network enviroment. We replaced this with two PHP scripts that run in a cron job. On the master server we run "mysqladmin flush-logs" to close the current binlog and start a new one, and then the PHP script gzips and gpgs the binlog and puts it somewhere where we can suck it across the Internet. The slave machine pulls down the file, un-gpgs and ungzips it, and loads the file. Since we run this every 5 minutes, we never worry about losing a lot of work in our database.
For as much as I'd like to see Dino win, I gotta say that I'm really glad that we don't have Ron Sims running our state. This guy is about as liberal as they get, so having Chris Gregoire run the state doesn't seem all that bad. I agree with the previous post that the politicians don't really care as long as they win, which is why there are so many citizen initiatives on every ballot that we have here. The voters have already pushed down taxes several times and when the liberal judges of our fair state overturned the ballot initiatives then the legislature saw the light and passed their own version of the same laws. I guess what I'm saying is that I'm not too worried about it right now.
(Before I get flamed by the Konqueror/Safari crowd, let me say that there are lots of DHTML rendering problems on these browsers when it comes to repainting backgrounds and tables on the fly. Since we test in IE and Mozilla/Firefox and Opera and it all works great, I guess the problems must be with those browsers. Our Mac customers don't mind installing Firefox, so why should we bother trying to find workarounds?)
Having spent more than my fair share of time doing bug triage, I know how hard it can be to prioritize which bugs get fixed and in what order, hence my comment about making the hard choices. I don't blame anyone for the number of bugs, nor do I blame them for not cleaning everything up right away. My criticism is that new feature development takes priority over bug cleanup (thereby creating even more bugs), and for poor communication on what's happening and why.