Try moving from VB6 to VB.NET. Most people I know decided that it would be little more pain to go and do c# instead.
I use OpenOffice.org, and I recently worked on a site with Office 2003, and so many of the buttons had moved or stupid things occurred that I didn't expect, that a transition to OOo would be no worse.
Thing is, they have sometimes entered a market with an existing product and made a better one.
Office 97 was the best office suite I'd seen.
The problem for Microsoft now seems to be that no-one is out there making new innovative PC based products, except in niche markets. They are just continuing the existing product lines, most of which seem to be sagging.
The real action is on the web. If someone builds a site or a web service, Microsoft have to compete head on. No-one thinks of it in terms of the desktop where all the MS stuff worked together.
The next big threat for Microsoft is applications getting off local servers and desktops and onto the web. Yes, I know that the net stations failed in 1998, but times is different now - bandwidth is cheap, 50% are on DSL and growing, and people are just familiar with having stuff on the net. If someone like Google came up with a high-quality on line calendar system (which included sharing) to add to Gmail, how many people would reconsider their use of Outlook/Exchange licenses?
Don't do anything for people who aren't friends, family or paying. Payment can come in a number of ways - share in a profit, cash or business barter.
They may make vague offers or promises, but frankly, if they are screwing you over, the minute you start asking for something paid, they'll just make themselves a stranger and find another freebie guy.
People who are decent people in business start paying straight away.
This doesn't mean that you shouldn't do nice things for your good customers, though. One customer of mine pays well for my coding, but I try and help them out with other bits of odd tech support for free.
I help friends and family of mine - not friends of friends. Not friends of family. Friends get help for free. Sometimes, they are generous back (like lunch, gift vouchers) but I don't expect it.
The rest are another matter. They pay. I don't even consider the "good recommendation" or "payback" thing, because I tried it once and found it to be lacking.
I once helped a realtor (who could definitely afford to pay), thinking I'd get some more work later. Gave him a CD with the downloaded driver he needed.
Response? Nothing. Not even a call to say it had worked. So, I phone the guy 2 days later, and he says that yes, it worked fine, and couldn't wait to get off the phone. No beer, no reference, no offer of barter. Nothing.
That taught me a good little lesson - Ask for the money. If someone is unwilling to pay, they're a cheapskate.
And never sell yourself cheap. It might only take 5 minutes to download a driver that you know they need, but they are paying for your expertise in knowing that.
The problem is that it's really hard for people to understand the price of custom work. They see that something can be bought from a company for $40 and somehow assume that an exact match for them from you will cost $40.
There's an old joke. Two men are in the jungle and a lion approaches them. The first man starts putting on his running shoes. The other says "you'll never outrun that lion", to which the first replies "I don't have to. I only have to outrun you".
In the context of Firefox security, the joke is that there are a whole lot of easier ways to attack someone's system than Firefox.
Let's consider what the steps are:-
Write an XPI to launch an attack.
Get it onto the Mozilla update site without anyone spotting it.
Hope that no-one spots its behaviour, even though the source code is in there.
Compare that with delivery by email of either a.exe or a.vbs, or putting something on a website that exploits someone and tricks them into downloading. It's a pretty crappy attack that's going to have a limited life. Even if it got through, people would be more wary after, and start checking the content of XPIs more thoroughly.
I can't believe how many people are banging on about blogging, like it's something completely amazing and new.
8 years ago, there were millions of personal web sites, full of people's top 10 trek episodes. Now, I see an entry on somewhere like Boing Boing or Slashdot, and within 2 hours, it's on someone else's blog. No comment about it, no enhancement, just copy it. I think most blogging will be dead in a year or 2.
If you are going to blog, try and follow the golden rule - make some content, or express an opinion that someone might want to read.
"These transistors may look fine, but let's see what happens when they're in as many radios as valves".
"These Japanese cars may look reliable now, but watch what happens when everyone's driving them".
"These quartz watches may look good, but when everyone's got them, we'll see if they're as good as a winding one".
Get the message? Some things are just better than others.
A documented XML-based format in a piece of open source software is much more likely to interoperate properly. If you can't read a document, there's a problem in one of two places. The program that produced it, or the program that's reading it, and YOU can validate it as being OK and trace that very quickly. Now, let's say you can't read a Word document. How do you know what's wrong. Do you know the definition? Can you send your document to Microsoft? You think they are going to take it apart for you and issue a patch to correct it?
As far as I know, OOo is already on a number of platforms. I've not heard of a single interoperability problem.
Someone once said the same thing about IBM. Now they say it about Microsoft (right now, you are right).
The thing that always shakes these things up is the next generation. What I'd ask is "what are the college graduates using?". If it's Linux, Microsoft should be very nervous, because these are the people in 10 years time setting up the next generation of small businesses, and fearless. If they've been using Linux at college and worked for them, they'll probably just keep on using it.
Yeah, all those major financial organisations use "toy" languages.
Who cares if you can or can't write an OS in Java. How many organisations want to write their own OS? You know any retailers who want to do that? Operating system development are a tiny part of the overall computer industry.
Java programmers are professionals, just like C++ programmers. They are often working in a different space - that of in-house application development. I don't know many companies using much C++ for in-house application development any more - it introduces risks and costs that aren't there with Java or.net.
Java and.net are the next step to the future. A future that will eventually be that businesses just graphically design their business models and rules, that are executed. There will always be a need for low-level programming for purposes like building operating systems and tools, but the desire of businesses is to have a shorter gap between requirements and delivery. That does not mean C++.
When I code, I want languages that do as much of the crappy stuff as possible. I want languages that get me to define business logic. I don't want to have to delete objects before they go out of scope. I don't want to have to pass a pointer to something, I just want to tell you to do it and you sort it out.
If you need the performance/flexibility of a language like C, it may be worth it, but often, such performance and flexibility are not required.
If you are running projects, what matters is cost - cost now and cost in the future. Memory leaks cost money in development, testing and client dissatisfaction. If a language does it for you, that's something else that you don't have to worry about and can make your system run better.
One rule of optimisation and efficiency is not to make it more efficient than it needs to be. If something needs to respond in 4 seconds and you are doing it in 3, stop working on it. If you are over 4 seconds, get optimising. In my experience of systems, a few key features of a system (the 10% that deal with the 90% of requests) are worth optimising. Stuff like table management is rarely worth optimising because it rarely gets done.
Another thing that is worth remembering is how cheap PCs are and how expensive programming time is. If a function is too slow and only used by 2 people, it may be cheaper to swap out their PCs than to change the software.
I run projects developing business database-based software. Stuff like java and.net suits me fine. Do I need to access hardware beyond what those languages give me? Nope. Ever likely to? Probably not, but there are ways to interface.
How critical is performance? It's as important as it needs to be. If a daily process has a 12 hour window to run in and takes 1 hour instead of 30 minutes, do I care? It's fast enough.
In addition, code performance matters less and less as machines get faster, memory goes up etc.
If you are involved in the 90% of the software industry that moves data around, the code bit of processing is taking a smaller and smaller slice of time compared to disk or network IO.
Concentrating on using safer, more productive languages and making code easier to understand is the direction we should all be aiming for.
We all have CD Roms for software as well. When I bought my first PC (a P75), Windows 95 came on CD and the first game I installed was a floppy based golf game.
Also a lot of people like to listen to their CDs while they are working/watch dumb movie clips on the internet.
When you first switch on the PC first time, it should lock down the network connections, so that on the net, it has 1 trusted connection by default - to Windows Update. Nothing in or out from anywhere else. Once Windows Update is complete, it can then go play on the net.
There are two levels of SEO work - one is just making your site better for a search engine to read - make it nice and compliant, that sort of thing. I'd count that as being OK and a good thing for the web overall.
The other way is to try and deliberately skew the results through link farms or jamming up blogs with your domain name.
I've met people promising people the world concerning where they'll be on Google's page, and all I can think is "and what happens when the algorithm changes?".
The problem in the UK is that we have a huge choice of water from UK and French sources that are spring water, which rightly or wrongly people judge to be better. You can't compete with water coming from romantic Sidcup.
Slashdot Mirror
Try moving from VB6 to VB.NET. Most people I know decided that it would be little more pain to go and do c# instead.
I use OpenOffice.org, and I recently worked on a site with Office 2003, and so many of the buttons had moved or stupid things occurred that I didn't expect, that a transition to OOo would be no worse.
Office 97 was the best office suite I'd seen.
The problem for Microsoft now seems to be that no-one is out there making new innovative PC based products, except in niche markets. They are just continuing the existing product lines, most of which seem to be sagging.
The real action is on the web. If someone builds a site or a web service, Microsoft have to compete head on. No-one thinks of it in terms of the desktop where all the MS stuff worked together.
The next big threat for Microsoft is applications getting off local servers and desktops and onto the web. Yes, I know that the net stations failed in 1998, but times is different now - bandwidth is cheap, 50% are on DSL and growing, and people are just familiar with having stuff on the net. If someone like Google came up with a high-quality on line calendar system (which included sharing) to add to Gmail, how many people would reconsider their use of Outlook/Exchange licenses?
Did all the money also all hit in the same year? In other words, if someone signed up for 3 years, is it all in 2004?
The ironic thing is that at one time, people used to set their clocks by the trains.
People were probably writing similar things when the fountain pen arrived (the death of the quill?).
They may make vague offers or promises, but frankly, if they are screwing you over, the minute you start asking for something paid, they'll just make themselves a stranger and find another freebie guy.
People who are decent people in business start paying straight away.
This doesn't mean that you shouldn't do nice things for your good customers, though. One customer of mine pays well for my coding, but I try and help them out with other bits of odd tech support for free.
I help friends and family of mine - not friends of friends. Not friends of family. Friends get help for free. Sometimes, they are generous back (like lunch, gift vouchers) but I don't expect it.
The rest are another matter. They pay. I don't even consider the "good recommendation" or "payback" thing, because I tried it once and found it to be lacking.
I once helped a realtor (who could definitely afford to pay), thinking I'd get some more work later. Gave him a CD with the downloaded driver he needed.
Response? Nothing. Not even a call to say it had worked. So, I phone the guy 2 days later, and he says that yes, it worked fine, and couldn't wait to get off the phone. No beer, no reference, no offer of barter. Nothing.
That taught me a good little lesson - Ask for the money. If someone is unwilling to pay, they're a cheapskate.
And never sell yourself cheap. It might only take 5 minutes to download a driver that you know they need, but they are paying for your expertise in knowing that.
The problem is that it's really hard for people to understand the price of custom work. They see that something can be bought from a company for $40 and somehow assume that an exact match for them from you will cost $40.
In the context of Firefox security, the joke is that there are a whole lot of easier ways to attack someone's system than Firefox.
Let's consider what the steps are:-
Write an XPI to launch an attack.
Get it onto the Mozilla update site without anyone spotting it.
Hope that no-one spots its behaviour, even though the source code is in there.
Compare that with delivery by email of either a .exe or a .vbs, or putting something on a website that exploits someone and tricks them into downloading. It's a pretty crappy attack that's going to have a limited life. Even if it got through, people would be more wary after, and start checking the content of XPIs more thoroughly.
8 years ago, there were millions of personal web sites, full of people's top 10 trek episodes. Now, I see an entry on somewhere like Boing Boing or Slashdot, and within 2 hours, it's on someone else's blog. No comment about it, no enhancement, just copy it. I think most blogging will be dead in a year or 2.
If you are going to blog, try and follow the golden rule - make some content, or express an opinion that someone might want to read.
"These Japanese cars may look reliable now, but watch what happens when everyone's driving them".
"These quartz watches may look good, but when everyone's got them, we'll see if they're as good as a winding one".
Get the message? Some things are just better than others.
A documented XML-based format in a piece of open source software is much more likely to interoperate properly. If you can't read a document, there's a problem in one of two places. The program that produced it, or the program that's reading it, and YOU can validate it as being OK and trace that very quickly. Now, let's say you can't read a Word document. How do you know what's wrong. Do you know the definition? Can you send your document to Microsoft? You think they are going to take it apart for you and issue a patch to correct it?
As far as I know, OOo is already on a number of platforms. I've not heard of a single interoperability problem.
The thing that always shakes these things up is the next generation. What I'd ask is "what are the college graduates using?". If it's Linux, Microsoft should be very nervous, because these are the people in 10 years time setting up the next generation of small businesses, and fearless. If they've been using Linux at college and worked for them, they'll probably just keep on using it.
The STL still allows memory leaks to occur, does it not?
Who cares if you can or can't write an OS in Java. How many organisations want to write their own OS? You know any retailers who want to do that? Operating system development are a tiny part of the overall computer industry.
Java programmers are professionals, just like C++ programmers. They are often working in a different space - that of in-house application development. I don't know many companies using much C++ for in-house application development any more - it introduces risks and costs that aren't there with Java or .net.
Java and .net are the next step to the future. A future that will eventually be that businesses just graphically design their business models and rules, that are executed. There will always be a need for low-level programming for purposes like building operating systems and tools, but the desire of businesses is to have a shorter gap between requirements and delivery. That does not mean C++.
When I code, I want languages that do as much of the crappy stuff as possible. I want languages that get me to define business logic. I don't want to have to delete objects before they go out of scope. I don't want to have to pass a pointer to something, I just want to tell you to do it and you sort it out.
If you are running projects, what matters is cost - cost now and cost in the future. Memory leaks cost money in development, testing and client dissatisfaction. If a language does it for you, that's something else that you don't have to worry about and can make your system run better.
One rule of optimisation and efficiency is not to make it more efficient than it needs to be. If something needs to respond in 4 seconds and you are doing it in 3, stop working on it. If you are over 4 seconds, get optimising. In my experience of systems, a few key features of a system (the 10% that deal with the 90% of requests) are worth optimising. Stuff like table management is rarely worth optimising because it rarely gets done.
Another thing that is worth remembering is how cheap PCs are and how expensive programming time is. If a function is too slow and only used by 2 people, it may be cheaper to swap out their PCs than to change the software.
I run projects developing business database-based software. Stuff like java and .net suits me fine. Do I need to access hardware beyond what those languages give me? Nope. Ever likely to? Probably not, but there are ways to interface.
How critical is performance? It's as important as it needs to be. If a daily process has a 12 hour window to run in and takes 1 hour instead of 30 minutes, do I care? It's fast enough.
If you are involved in the 90% of the software industry that moves data around, the code bit of processing is taking a smaller and smaller slice of time compared to disk or network IO.
Concentrating on using safer, more productive languages and making code easier to understand is the direction we should all be aiming for.
Also a lot of people like to listen to their CDs while they are working/watch dumb movie clips on the internet.
When you first switch on the PC first time, it should lock down the network connections, so that on the net, it has 1 trusted connection by default - to Windows Update. Nothing in or out from anywhere else. Once Windows Update is complete, it can then go play on the net.
It's because the expectation has been lowered. People assume that all this stuff is normal, probably because that's what all their friends get too.
The other way is to try and deliberately skew the results through link farms or jamming up blogs with your domain name.
I've met people promising people the world concerning where they'll be on Google's page, and all I can think is "and what happens when the algorithm changes?".
Shuttle (who make the lovely barebones boxes) are working seriously with Mandrake on making tested hardware.
Seriously, how many people need stuff in Office that isn't in OpenOffice.org?
The problem in the UK is that we have a huge choice of water from UK and French sources that are spring water, which rightly or wrongly people judge to be better. You can't compete with water coming from romantic Sidcup.