Working from home only for developers ?

[juststarting]

the last few php contracts i did were 350 per day in london, and 250 per day outside.

LAMP on the way out? thats an odd statement to make - as the contracts i was involved with were moving in the opposite way - from dot net to LAMP.

maybe its a credit crunch effect - costs are trumping code snobbery...


good one to check out juststarting is SugarCRM. no need to code it - just get to know how it works. brush up on business analysis... theres a demand out there for biz analysts/project managers who can plan and design a sugarcrm site. and since its web based , a lot of this can be done remotely.

plus you can download it for free and have a play.

CRM is something I planned for future ... I've been doing Sharepoint infrastructure and have very good infrastructure / operations knowledge in IIS / Sharepoint and very sound ASP. .Net basics. .

I haven't heard of SugarCRM but for the last 10 years I was sticking to Microsoft. It started with their big market , then their great support and documentation , I always had a reason to stay Microsoft. Hence it is natural for me to look into Dynamics , their CRM. Last thing I heard is that it has gone Web Based ..

[rephlex]

I don't hate PHP but wouldn't learning it as one of your first languages encourage exactly the kind of bad practices that give PHP developers a bad name?

i'm aware of that - but the chap said he was a non-programmer. and considering that php has a shallow learning curve, it might be an option to get stuck into it , considering the amount of php driven sites out there. especially when it comes to working from home.

and because of the shallow learning curve, it would give him an confidence boost, which is no bad thing. its a useful add-on to have to a portfolio of skills. mysql would be another one. not expert level - but enough to get by.

if i had my "career advice" hat on to a geek who really wanted to be a coder full time, it would be dot Net or Java for the money/career , with python getting a look in for those Google engineering jobs.

[rephlex]

CRM is something I planned for future ... I've been doing Sharepoint infrastructure and have very good infrastructure / operations knowledge in IIS / Sharepoint and very sound ASP. .Net basics. .

I haven't heard of SugarCRM but for the last 10 years I was sticking to Microsoft. It started with their big market , then their great support and documentation , I always had a reason to stay Microsoft. Hence it is natural for me to look into Dynamics , their CRM. Last thing I heard is that it has gone Web Based ..

it has - i've used it and it sucks. but worth sticking with -i'm sure microsoft will improve it on further releases.

but keep an eye out for Sugar... its gaining traction out there as folks start tightening the belts and start looking at more cost effective solutions.

very active community as well.

http://www.sugarcrm.com

[Jog On]

I'm making my first foray into coding via PHP/Mysql

It makes sense for my plan B to do that bit on my own which I'm enjoying in a 'why didn't I do this 9 years ago' kind of way....

Ho hum..

[NickFitz]

If you have the aptitude, getting your head around the basics of programming isn't that hard. Given said aptitude, it doesn't matter too much what language you start out with - an easy-to-grasp one like PHP isn't a bad start, although you should avoid the mistake of thinking that a knowledge of a set of libraries (in which most languages are very rich nowadays) is the same as an understanding of programming.

For example, you might have an excellent understanding of the libraries of APIs that come for free with PHP, Java, or any .NET language.

But if, for example, you can't recognise that the specific problem with which you are presented belongs to the set of NP-Complete problems, and might therefore be better approached (for the problem set with which it has to deal in reality, as opposed to the smaller problem set in your development environment) by using (for the sake of argument) a heuristic algorithm, rather than some more obvious but unscaleable algorithm, then you're not really there.

Even leaving aside such questions of the order of algorithms and their scaleability, take a simple problem:

There are twenty rooms, five of which are locked. The five keys for said locks must be distributed randomly among these rooms. However, it must be possible to unlock every room. So, for example, if the key to room A is in room B, the key to room B cannot be in room A. If the key to room A is in room B, and the key to room B is in room C, the key to room C cannot be in room A. And so on...

This problem requires understanding a variety of issues relating to graphs of dependencies. As stated, it is quite easily solved by the appropriate use of a recursive algorithm with backtracking as necessary. However, I am not aware of any library function in any framework that provides a direct solution. Even if there were such a library function, you would need to know enough to recognise it as providing the solution for the problem as described, for the documentation would describe it in more abstract terms than does the problem specification.

I solved it, and implemented it in 68000 assembler, in about 45 minutes (including some other details), down the pub with a pen and paper in 1989. But I'd had a lot of experience of programming in many languages by then... and nowadays it'd probably take more like ten minutes. The point is, knowing a language and its associated toolkit isn't the same as applying them to a specific problem description - knowing the tools only helps if you know when and how to apply them, and when and how to build new tools for problems that weren't foreseen by the toolkit's creators.

On the odd occasion that I've been involved in interviewing, I've always been deeply suspicious of people who claim to be able to pick up a new framework quickly. A new language... well, if you have a good understanding of programming, then you can learn the basics of a new language very quickly, although it'll take you a while to become skilled at it (I'm still getting my head around Python - I've used it professionally, but I'm still not convinced that I understand it well). Modern frameworks (and as I said, I include things like the Java APIs, the .NET framework, and the enormous number of functions in PHP among these) take a lot of experience. This is one reason that I no longer claim expertise with ASP.NET: I used to know it reasonably well, but it's grown since then, and I've had no practical experience with its bigger, better form.

An aside to TheBigYinJames: WTF are you on? The only change I've seen in relation to LAMP is that more people are choosing Python as the P... Linux servers are still dominant, Apache web servers are still dominant, mySQL database servers are still dominant... well, for anybody building a public-facing web application anyway. (YouTube and much else at Google is written in Python, and Yahoo! runs on PHP, FWTW )

[TheBigYinJames]

LAMP on the way out? thats an odd statement to make - as the contracts i was involved with were moving in the opposite way - from dot net to LAMP.

I'm sure that's an effect of you being a LAMPer. I'm not having a go at LAMP, I used to do it myself, but the bottom fell out of that market about 4 or 5 years ago, so I switched to dotNet to keep my rates up.

Since I took the three Ps (Perl/PHP/Python) off my CV, I've noticed that because it's now a niche skill, I occassionally get offered megabucks to do it - obviously some places are heavily reliant on it and can't get off the train. This doesn't mean that it's suddenly a highly sought-after skill, just that nobody els is doing it any more. I've resisted so far, because I don't want to lose my .Net current experience, which was hard won.

But if you look at the number and rates of jobs going, LAMP is way down there compared with the big two of Java and .Net, so I was trying to steer them in a direction where the market appears to be holding steady, rather than LAMP which is becoming niche.

[TheBigYinJames]

An aside to TheBigYinJames: WTF are you on? The only change I've seen in relation to LAMP is that more people are choosing Python as the P... Linux servers are still dominant, Apache web servers are still dominant, mySQL database servers are still dominant... well, for anybody building a public-facing web application anyway. (YouTube and much else at Google is written in Python, and Yahoo! runs on PHP, FWTW )

Well I haven't looked into it in massive detail, just doing a rough compare on the number of jobs going on jobserve. If it's still going healthy then more power to its collective elbows, perhaps it's more of a word of mouth sector.

Also I know the individual parts of of LAMP are still used extensively separately, but LAMP as a unified concept for 20 years of development I don't think is the strongest skillset to set yourself up in from the off, if you have a choice.

[tim123]

take a simple problem:

There are twenty rooms, five of which are locked. The five keys for said locks must be distributed randomly among these rooms. However, it must be possible to unlock every room. So, for example, if the key to room A is in room B, the key to room B cannot be in room A. If the key to room A is in room B, and the key to room B is in room C, the key to room C cannot be in room A. And so on...

What's this example got to do with "real world"programming.

Most programming problems are

"user clicks button" (real or virtual)

"Read inputs and calculate result"

"Give choice of responses to user".

"process response"

Learning how to do these things, in the environment you are using, is 95% of the task.

The example above is just theoretical nonsense

tim

[EternalOptimist]

What's this example got to do with "real world"programming.

Most programming problems are

"user clicks button" (real or virtual)

"Read inputs and calculate result"

"Give choice of responses to user".

"process response"

Learning how to do these things, in the environment you are using, is 95% of the task.

The example above is just theoretical nonsense

tim

He is saying that there are quite a few skills required to be a developer, and he feels that programming and problem solving are the main two.


On a point of interest, I wonder how many skills overall a developer needs compared say, with a DBA ?

[DimPrawn]

What's this example got to do with "real world"programming.

Most programming problems are

"user clicks button" (real or virtual)

"Read inputs and calculate result"

"Give choice of responses to user".

"process response"

Learning how to do these things, in the environment you are using, is 95% of the task.

The example above is just theoretical nonsense

tim

I think NickFitz has a good example there when testing developers. Anyone who can solve Nicks logic puzzle is kept well away from the management, users and anyone else living in the real world.

HTH