In the US you'd probably expect this to be privatised and profit making, but unlike most countries, the mapping data from the national agencies is freely available. By comparison, the UK Ordnance Survey maps are not free, even though they're paid by the taxpayers.
OSM has attempted to use the free Tiger data set but ran into a hitch with the import and has been trying ever since to get the import working. I'm not sure why this is taking so long, other than perhaps there are few people interested in mapping the US. As far as I can tell, the people who've got involved with OSM are concentrated in the UK.
But don't look initially at the completeness. Instead take a look at the design and tech. It's pretty easy to see that they've got most stuff right and just need people getting the data added. Take a look around University of Manchester area - there was a mapping event there and there's a good section of complete maps that looks pretty good.
Adding up and dividing by the number gives an average - the mean average. The median is another average, as is mode. Average isn't a particularly technical term, it's not referring to any specific algorithm.
The mean average salary probably isn't that useful - I'd be interested to know the modal average - what the typical person earns. E.g. I know here in the UK the (mean) average salary is about £24,000 but the majority of people I know earn less than that - the mean is skewed quite a lot by the super rich - Richard Branson alone probably contributes hundreds of pounds to the mean average. I'd guess the modal average is under £20,000.
Dual licensed code is just pick and choose - the licences don't have to be compatible. If there's a GPL licence on it, it's GPL compatible because you don't HAVE to follow the other licence(s). Just to clear up my own POV here, I think that technically the Linux guys haven't done anything wrong by removing the BSD licence. The files had a choice of two licences - they made a choice.
The thing that irks me about it is just that there was no need to make that choice. Just leave the two licences intact. The BSD licence does allow downstream people to be gits, the GPL terms prevent them from it. IMO, in this case, GPL developers are doing to BSD developers exactly what their licence aims to prevent other developers doing to them. Maybe you see that as a plus point for the GPL, but I see it as hypocritical.
The BSD licence includes "Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer." If you're distributing source code, you have to follow that. I believe the complaint was that it was removed.
With regards to BSD software, it being re-licenced solely as GPL has the same effect as being used and modified in commercial software - improvements can't then be copied back to the BSD version. So from a technical standpoint it's only equivalent to what the licence is allowing people to do whenever they like. But just from an ethical point of view, it just feels wrong for a free software advocate to take the software out of the hands of the BSD crowd.
With BSD licenced software, I can work with the same code both as a proprietary developer, and when working on open source stuff in my spare time. Although I can't automatically copy my proprietary code back to BSD (though I suspect it would be allowed if I asked to do it), I'd never consider removing a BSD licence if I was working on open source stuff on another licence.
Unfortunately all your counterpoints are pretty arguable also. There was no argument for stopping transatlantic flights - only pointing out the implications of taking one. The cooling effects of contrails are not yet understood - studies have differed in opinion as to whether they warm or cool the earth - currently Wikipedia says they have a warming effect which is my recollection of the balance of opinion. And as for causing deaths, events made more likely by climate change are already causing deaths.
The science of climate change is no longer under debate in peer-reviewed articles. The predictions of global warming are built on firm theory and well understood mechanisms. They are well supported by measurements of the increasing global temperature. It's pretty clear that the ice caps are going to melt and raise sea levels, at the current rate.
A lot of people live in fixed coastal cities. When the sea levels rise, they'll build dams to keep the cities safe. I'd certainly not bet that every one of these dams will hold. There's plenty of other ways that people can die as a result of climate change anyway.
Didn't Bell Labs and university researchers come up with Lithium Ion technology. Sony were first to market commercially, but I've never seen anything crediting them with the invention of the technology.
The Passat is a car I have driven frequently as a hire car. I actually think it's a pretty rubbish car - no one seems to want one as I always get it as a free 'upgrade' from the class of car we're allowed to hire (which includes the much better VW Golf, and the even better than that Honda Civic).
The 2.0 TDI Passat gets over 50 mpg on the motorway, almost exactly on the quoted figure. I've never driven one around town much.
To get better MPG, people are going to have to get used to smaller cars, and many modern small cars are really well designed. The Toyota Yaris is an excellent example. I'm 6 feet tall and 19 stones - big even by US standards, and the Yaris doesn't feel cramped to me, although I do put the seat all the way back. I only got a Ford Focus in the end because I was buying second-hand, and the ridiculous depreciation on any Ford makes them a bargain vs second hand Toyotas.
Although I don't agree with whaling, I feel I should at least point out that the IWC was a whaling industry organisation that was subverted into a conservation group. This happened as a group of big countries recruited a lot of the smaller countries (with again, no whaling interests of their own) to join the IWC and vote to ban whaling.
So this ruthless vote buying process is hardly without precedent...
The Jetta certainly isn't one of the 80 MPG cars - it's pretty big by European standards - there are 3 smaller models in the UK line-up. It depends on exactly which model you got and where you were driving I guess. Automatics are significantly lower efficiency than manuals - in fact for the smallest diesel engined Jetta, with automatic transmission, driving in town, the quoted fuel economy is 36 mpg.
There is a VW Polo model that does 57mpg urban and over 88mpg extra-urban. It's one of the 1.4 diesels and there are quite a few similar models with similar fuel economy.
This black puff of smoke idea, I'm not too sure where you get this idea - I rarely see this. I've just got a diesel myself, 5 year old Ford Focus, so fairly average in terms of emissions. There is a very small amount of soot when the accelerator is genuinely floored, but even driving aggressively it's pretty rare and pointless to floor it, and I'm not sure the tiny amount of smoke would be noticed by any other drivers. Only poorly maintained cars or really old ones emit a significant amount of soot.
It's not about code reading like English, but more about it being easy for another coder to work with. One of my co-workers is a brilliant coder but poor at spelling. While customising a standard project component he had written I was thoroughly impressed by how well thought out and tightly coded everything was.
Unfortunately when I tried to compile my customised version, there were hundreds of errors, the majority of which were where I had spelled words correctly. Mostly I fixed these by changing the spellings in my code, but there were a couple of places where I'd somehow ended up with compilable code that was using the wrong variable, resulting in run-time crashes.
The difficulty is, any spell checker would have to realise that TcpRcv, TcpRcvr, probably even TcpRecvr are valid abbreviations, but TcpReciever is a spelling mistake. I think rather than looking at whole words, it would probably have to look for problematic sequences of letters, such as 'cie' which should almost always be 'cei'.
That's a load of rubbish. Because of the compression (MP3 or AAC) the bits being wrong lead very quickly to extremely noticeable defects (clicking and popping noises) in the tracks. MP3 is quite robust and remains playable, but AVI files tend to get completely unplayable with a few bits changed in the wrong places. I've never had this problem or any data-loss problems on my iPod - I've had it once when burning a backup CD on a rewritable disc which must have been used too many times.
If you've got those problems with your iPod it's a defective unit. I can believe that iPod units may have a failure rate of 1 in 100 (which is actually very low compared to most modern gadgets), but on a properly functioning iPod, this memory miss rate isn't true.
Only the first access causes the resize - after that, the resized image is cached. It's just usually the Google spider gets there before anything else, normally within a couple of hours of new photos being uploaded.
A great point. I/O to discs and networks has a few cases where the standard performance is poor. I've recently got a copy of Windows Vista and this seems to be even worse than I've seen before - performance during file copies across the network is horrible.
I've got a media server and HTPC networked up, and I'd really like to be able to be watching one film from the server, while copying a new DVD back to the server. Right now, setting the player priority to high has no effect on the network transfer and I can't get a watchable film. Network I/O is always going to have uncontrollable outside influences, but in this case at least, the OS could sort things out.
Also I run a photos website on a Linux box. Occasionally the Google spider would be indexing the site and cause lots of photos to be resized. This uses some CPU but the major problem is disc I/O. I'd love to be able to just set Apache processes to low I/O priority so this wouldn't affect performance for other stuff too adversely.
Haven't Playstations always been the same? Sony isn't a software company and can't provide as much support as MS have done with the Xbox 360. Considering also the year lead of the 360, it's fairly safe to assume the PS3 would be about 18 months behind on software. Developers from all corners have been complaining of the difficulty or working with the PS3, which is the same as was true for PS1 and PS2.
The Xbox came with standard Direct-X programming tools that PC developers were already familiar with. On the Xbox, first gen games looked pretty similar to what was coming out later on. On PS2 by comparison, the bar was raised every year - it still is even now. E.g. Red Faction looked great in 2001, but it's rubbish by the standards of God of War 2.
Unfortunately everyone concentrates on getting the graphics bit right for the first year. In that sense the Wii did a great favour to it's buyers by pretty much being a statement that graphics weren't important - developers from the start just made some fun games. PS3 developers feel they at least have to outdo PS2 and Xbox graphics before they can look at gameplay. It'll work out over the next year.
Considering how many PS2 games were already using 2 or 3 DVDs, using higher-capacity media seems to make perfect sense for the next-gen console - it seems like a decent decision to make Blu-Ray a feature for that reason. For what it's worth the Blu-Ray player in the PS3 is excellent. I've got mine hooked up with HDMI to a 1080p TV and I've been watching the Planet Earth series - the quality is fantastic.
I'd have thought there was quite a lot of overlap between games enthusiasts and film / home theatre enthusiasts. Even more so for those who like their games beautifully rendered in high def goodness.
I think you've completely missed the point of the WTO - enabling free trade. If you allow a service or product, you have to allow companies (or individuals) from any country to compete to provide that.
The big US gambling companies run casinos in other countries and online making money from those overseas markets, but the US is effectively saying that other countries companies can't have access to the US market. That is clearly not free trade, and stopping such practises is exactly what the WTO is about (amongst other things).
Of course, if you create laws that forbid certain products - alcohol (Muslim coutries), speed camera detectors (Australia, I think) etc. - then that's a different matter, and no-one can sell them.
The cost of blowing up a rocket is much lower if you've not had to produce thousands of pages of documents though. It obviously is quite a lot of hardware, but I don't really believe the actual hardware costs the tens of millions commonly quoted even for small rockets.
In the UK, Blu-Ray shelf space seems to be significantly higher than HD-DVD (at HMV, Tesco and Asda at least). Not sure what's happenning on sales numbers but I'd expect retail shelf space to be following the sales figures.
Have a look at the cost to vendors of accepting card payments, and you'll probably change your mind about whether they like you. For typical small merchants, there is a 2.5% charge on the purchase cost which the vendor pays to the credit card company. Plenty of stores will pass this directly on to the customer, especially price-sensitive online shops. No doubt the Wal-Marts are only paying closer to 0.5% due to their purchasing power, but it's still quite a bit of money.
For your $50K of charges, your credit card company will have been paid ~ $1000 in fees. The breakdown of that between all the companies involved (Bank Of America, Visa or Mastercard etc) I'm not so sure of. Unless you're constantly ringing up customer services it's fairly certain they will be making a nice profit from your custom.
Actually, you've got it the wrong way round here. GDP growth is pretty much exponential, just on the approximate rate of 2-3% a year. Average businesses can experience exponential growth, just maybe not 50% a year.
This sounds like the right sort of idea - making the earliest stage of the process quicker and cheaper. I think another very worthwhile idea would be that as soon as any proceedings are filed on any patent claim, the patent in question should immediately be re-submitted for more thorough examination by the patent office. If the re-examination fails, the case is thrown out before it's begun.
Too many patents mean that there isn't enough resources to check each one as well as would be desired, but for the small proportion on which lawsuits are brought, a much better check could be carried out before-hand. This also means that patent trolls buying up the broad patents would in many cases be buying worthless pieces of paper that would be invalidated before they caused havoc.
I think the idea of getting things right the first time isn't such a big deal. The problem is once you've reported it as done if you realise it's not right, deadlines will pressure you into going onto something else - a degree of experience and conviction is required to say it needs re-doing. Generally, the best people will not only see things that are wrong, but will also fix the things that are just not right. But this is a bit of a catch-22. Most commercial software is done to timescales and so only recognised experts can get the freedom to fix things that aren't wrong - but half the time you can't tell who the experts are until they've rewritten stuff you hadn't realised needed rewriting.
The balance to strike is between getting things right and getting things done, and this is the biggest area that 'experts' tend to be best at and one where experience is valuable. Some bits of the code need to be done right so the rest can simply fall into place. I work on custom one-off pieces of software that are loosely based around core modules. These are frequently re-used so they need to be done properly. Most of the application specific stuff just needs to be done. Getting the balance just right on all the different bits of the system is what really makes the huge differences to productivity between developers.
I've always been pretty good at maths and can understand it with a little effort. But I think maths is pretty similar to learning a second language - once you get used to it, you stop translating it to English and the formula IS the maths. I never got good enough at a foreign language that I could stop translating to/from English.
Having written all this, I wonder if for many people maths is just a lot like doing a French lesson where the teacher jumps immediately to hosting the class in French, and explaining what you're supposed to be doing in French. Back when I was at school, I tended to get a bit lost in these sort of lessons. It almost seems odd that it's not often the same people who can do languages that can do maths
With XAMPP it really is pretty simple. I've just reinstalled my PC and used XAMPP once again. It's a quick graphic installer that gives you an immediate working LAMP stack. All that was left to do was set the MySQL data directory and Apache document root to the copied folder from my previous install and all the database apps were back up and running.
Slashdot Mirror
In the US you'd probably expect this to be privatised and profit making, but unlike most countries, the mapping data from the national agencies is freely available. By comparison, the UK Ordnance Survey maps are not free, even though they're paid by the taxpayers.
OSM has attempted to use the free Tiger data set but ran into a hitch with the import and has been trying ever since to get the import working. I'm not sure why this is taking so long, other than perhaps there are few people interested in mapping the US. As far as I can tell, the people who've got involved with OSM are concentrated in the UK.
But don't look initially at the completeness. Instead take a look at the design and tech. It's pretty easy to see that they've got most stuff right and just need people getting the data added. Take a look around University of Manchester area - there was a mapping event there and there's a good section of complete maps that looks pretty good.
Adding up and dividing by the number gives an average - the mean average. The median is another average, as is mode. Average isn't a particularly technical term, it's not referring to any specific algorithm.
The mean average salary probably isn't that useful - I'd be interested to know the modal average - what the typical person earns. E.g. I know here in the UK the (mean) average salary is about £24,000 but the majority of people I know earn less than that - the mean is skewed quite a lot by the super rich - Richard Branson alone probably contributes hundreds of pounds to the mean average. I'd guess the modal average is under £20,000.
Dual licensed code is just pick and choose - the licences don't have to be compatible. If there's a GPL licence on it, it's GPL compatible because you don't HAVE to follow the other licence(s). Just to clear up my own POV here, I think that technically the Linux guys haven't done anything wrong by removing the BSD licence. The files had a choice of two licences - they made a choice.
The thing that irks me about it is just that there was no need to make that choice. Just leave the two licences intact. The BSD licence does allow downstream people to be gits, the GPL terms prevent them from it. IMO, in this case, GPL developers are doing to BSD developers exactly what their licence aims to prevent other developers doing to them. Maybe you see that as a plus point for the GPL, but I see it as hypocritical.
The BSD licence includes "Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer." If you're distributing source code, you have to follow that. I believe the complaint was that it was removed.
With regards to BSD software, it being re-licenced solely as GPL has the same effect as being used and modified in commercial software - improvements can't then be copied back to the BSD version. So from a technical standpoint it's only equivalent to what the licence is allowing people to do whenever they like. But just from an ethical point of view, it just feels wrong for a free software advocate to take the software out of the hands of the BSD crowd.
With BSD licenced software, I can work with the same code both as a proprietary developer, and when working on open source stuff in my spare time. Although I can't automatically copy my proprietary code back to BSD (though I suspect it would be allowed if I asked to do it), I'd never consider removing a BSD licence if I was working on open source stuff on another licence.
Unfortunately all your counterpoints are pretty arguable also. There was no argument for stopping transatlantic flights - only pointing out the implications of taking one. The cooling effects of contrails are not yet understood - studies have differed in opinion as to whether they warm or cool the earth - currently Wikipedia says they have a warming effect which is my recollection of the balance of opinion. And as for causing deaths, events made more likely by climate change are already causing deaths.
The science of climate change is no longer under debate in peer-reviewed articles. The predictions of global warming are built on firm theory and well understood mechanisms. They are well supported by measurements of the increasing global temperature. It's pretty clear that the ice caps are going to melt and raise sea levels, at the current rate.
A lot of people live in fixed coastal cities. When the sea levels rise, they'll build dams to keep the cities safe. I'd certainly not bet that every one of these dams will hold. There's plenty of other ways that people can die as a result of climate change anyway.
Didn't Bell Labs and university researchers come up with Lithium Ion technology. Sony were first to market commercially, but I've never seen anything crediting them with the invention of the technology.
The Passat is a car I have driven frequently as a hire car. I actually think it's a pretty rubbish car - no one seems to want one as I always get it as a free 'upgrade' from the class of car we're allowed to hire (which includes the much better VW Golf, and the even better than that Honda Civic).
The 2.0 TDI Passat gets over 50 mpg on the motorway, almost exactly on the quoted figure. I've never driven one around town much.
To get better MPG, people are going to have to get used to smaller cars, and many modern small cars are really well designed. The Toyota Yaris is an excellent example. I'm 6 feet tall and 19 stones - big even by US standards, and the Yaris doesn't feel cramped to me, although I do put the seat all the way back. I only got a Ford Focus in the end because I was buying second-hand, and the ridiculous depreciation on any Ford makes them a bargain vs second hand Toyotas.
Although I don't agree with whaling, I feel I should at least point out that the IWC was a whaling industry organisation that was subverted into a conservation group. This happened as a group of big countries recruited a lot of the smaller countries (with again, no whaling interests of their own) to join the IWC and vote to ban whaling.
...
So this ruthless vote buying process is hardly without precedent
The Jetta certainly isn't one of the 80 MPG cars - it's pretty big by European standards - there are 3 smaller models in the UK line-up. It depends on exactly which model you got and where you were driving I guess. Automatics are significantly lower efficiency than manuals - in fact for the smallest diesel engined Jetta, with automatic transmission, driving in town, the quoted fuel economy is 36 mpg.
There is a VW Polo model that does 57mpg urban and over 88mpg extra-urban. It's one of the 1.4 diesels and there are quite a few similar models with similar fuel economy.
This black puff of smoke idea, I'm not too sure where you get this idea - I rarely see this. I've just got a diesel myself, 5 year old Ford Focus, so fairly average in terms of emissions. There is a very small amount of soot when the accelerator is genuinely floored, but even driving aggressively it's pretty rare and pointless to floor it, and I'm not sure the tiny amount of smoke would be noticed by any other drivers. Only poorly maintained cars or really old ones emit a significant amount of soot.
It's not about code reading like English, but more about it being easy for another coder to work with. One of my co-workers is a brilliant coder but poor at spelling. While customising a standard project component he had written I was thoroughly impressed by how well thought out and tightly coded everything was.
Unfortunately when I tried to compile my customised version, there were hundreds of errors, the majority of which were where I had spelled words correctly. Mostly I fixed these by changing the spellings in my code, but there were a couple of places where I'd somehow ended up with compilable code that was using the wrong variable, resulting in run-time crashes.
The difficulty is, any spell checker would have to realise that TcpRcv, TcpRcvr, probably even TcpRecvr are valid abbreviations, but TcpReciever is a spelling mistake. I think rather than looking at whole words, it would probably have to look for problematic sequences of letters, such as 'cie' which should almost always be 'cei'.
That's a load of rubbish. Because of the compression (MP3 or AAC) the bits being wrong lead very quickly to extremely noticeable defects (clicking and popping noises) in the tracks. MP3 is quite robust and remains playable, but AVI files tend to get completely unplayable with a few bits changed in the wrong places. I've never had this problem or any data-loss problems on my iPod - I've had it once when burning a backup CD on a rewritable disc which must have been used too many times.
If you've got those problems with your iPod it's a defective unit. I can believe that iPod units may have a failure rate of 1 in 100 (which is actually very low compared to most modern gadgets), but on a properly functioning iPod, this memory miss rate isn't true.
Only the first access causes the resize - after that, the resized image is cached. It's just usually the Google spider gets there before anything else, normally within a couple of hours of new photos being uploaded.
A great point. I/O to discs and networks has a few cases where the standard performance is poor. I've recently got a copy of Windows Vista and this seems to be even worse than I've seen before - performance during file copies across the network is horrible.
I've got a media server and HTPC networked up, and I'd really like to be able to be watching one film from the server, while copying a new DVD back to the server. Right now, setting the player priority to high has no effect on the network transfer and I can't get a watchable film. Network I/O is always going to have uncontrollable outside influences, but in this case at least, the OS could sort things out.
Also I run a photos website on a Linux box. Occasionally the Google spider would be indexing the site and cause lots of photos to be resized. This uses some CPU but the major problem is disc I/O. I'd love to be able to just set Apache processes to low I/O priority so this wouldn't affect performance for other stuff too adversely.
Haven't Playstations always been the same? Sony isn't a software company and can't provide as much support as MS have done with the Xbox 360. Considering also the year lead of the 360, it's fairly safe to assume the PS3 would be about 18 months behind on software. Developers from all corners have been complaining of the difficulty or working with the PS3, which is the same as was true for PS1 and PS2.
The Xbox came with standard Direct-X programming tools that PC developers were already familiar with. On the Xbox, first gen games looked pretty similar to what was coming out later on. On PS2 by comparison, the bar was raised every year - it still is even now. E.g. Red Faction looked great in 2001, but it's rubbish by the standards of God of War 2.
Unfortunately everyone concentrates on getting the graphics bit right for the first year. In that sense the Wii did a great favour to it's buyers by pretty much being a statement that graphics weren't important - developers from the start just made some fun games. PS3 developers feel they at least have to outdo PS2 and Xbox graphics before they can look at gameplay. It'll work out over the next year.
Considering how many PS2 games were already using 2 or 3 DVDs, using higher-capacity media seems to make perfect sense for the next-gen console - it seems like a decent decision to make Blu-Ray a feature for that reason. For what it's worth the Blu-Ray player in the PS3 is excellent. I've got mine hooked up with HDMI to a 1080p TV and I've been watching the Planet Earth series - the quality is fantastic.
I'd have thought there was quite a lot of overlap between games enthusiasts and film / home theatre enthusiasts. Even more so for those who like their games beautifully rendered in high def goodness.
I think you've completely missed the point of the WTO - enabling free trade. If you allow a service or product, you have to allow companies (or individuals) from any country to compete to provide that.
The big US gambling companies run casinos in other countries and online making money from those overseas markets, but the US is effectively saying that other countries companies can't have access to the US market. That is clearly not free trade, and stopping such practises is exactly what the WTO is about (amongst other things).
Of course, if you create laws that forbid certain products - alcohol (Muslim coutries), speed camera detectors (Australia, I think) etc. - then that's a different matter, and no-one can sell them.
The cost of blowing up a rocket is much lower if you've not had to produce thousands of pages of documents though. It obviously is quite a lot of hardware, but I don't really believe the actual hardware costs the tens of millions commonly quoted even for small rockets.
In the UK, Blu-Ray shelf space seems to be significantly higher than HD-DVD (at HMV, Tesco and Asda at least). Not sure what's happenning on sales numbers but I'd expect retail shelf space to be following the sales figures.
Have a look at the cost to vendors of accepting card payments, and you'll probably change your mind about whether they like you. For typical small merchants, there is a 2.5% charge on the purchase cost which the vendor pays to the credit card company. Plenty of stores will pass this directly on to the customer, especially price-sensitive online shops. No doubt the Wal-Marts are only paying closer to 0.5% due to their purchasing power, but it's still quite a bit of money.
For your $50K of charges, your credit card company will have been paid ~ $1000 in fees. The breakdown of that between all the companies involved (Bank Of America, Visa or Mastercard etc) I'm not so sure of. Unless you're constantly ringing up customer services it's fairly certain they will be making a nice profit from your custom.
Actually, you've got it the wrong way round here. GDP growth is pretty much exponential, just on the approximate rate of 2-3% a year. Average businesses can experience exponential growth, just maybe not 50% a year.
This sounds like the right sort of idea - making the earliest stage of the process quicker and cheaper. I think another very worthwhile idea would be that as soon as any proceedings are filed on any patent claim, the patent in question should immediately be re-submitted for more thorough examination by the patent office. If the re-examination fails, the case is thrown out before it's begun.
Too many patents mean that there isn't enough resources to check each one as well as would be desired, but for the small proportion on which lawsuits are brought, a much better check could be carried out before-hand. This also means that patent trolls buying up the broad patents would in many cases be buying worthless pieces of paper that would be invalidated before they caused havoc.
I think the idea of getting things right the first time isn't such a big deal. The problem is once you've reported it as done if you realise it's not right, deadlines will pressure you into going onto something else - a degree of experience and conviction is required to say it needs re-doing. Generally, the best people will not only see things that are wrong, but will also fix the things that are just not right. But this is a bit of a catch-22. Most commercial software is done to timescales and so only recognised experts can get the freedom to fix things that aren't wrong - but half the time you can't tell who the experts are until they've rewritten stuff you hadn't realised needed rewriting.
The balance to strike is between getting things right and getting things done, and this is the biggest area that 'experts' tend to be best at and one where experience is valuable. Some bits of the code need to be done right so the rest can simply fall into place. I work on custom one-off pieces of software that are loosely based around core modules. These are frequently re-used so they need to be done properly. Most of the application specific stuff just needs to be done. Getting the balance just right on all the different bits of the system is what really makes the huge differences to productivity between developers.
I've always been pretty good at maths and can understand it with a little effort. But I think maths is pretty similar to learning a second language - once you get used to it, you stop translating it to English and the formula IS the maths. I never got good enough at a foreign language that I could stop translating to/from English.
Having written all this, I wonder if for many people maths is just a lot like doing a French lesson where the teacher jumps immediately to hosting the class in French, and explaining what you're supposed to be doing in French. Back when I was at school, I tended to get a bit lost in these sort of lessons. It almost seems odd that it's not often the same people who can do languages that can do maths
With XAMPP it really is pretty simple. I've just reinstalled my PC and used XAMPP once again. It's a quick graphic installer that gives you an immediate working LAMP stack. All that was left to do was set the MySQL data directory and Apache document root to the copied folder from my previous install and all the database apps were back up and running.