Re: Learning process
From: Paul F. Johnson (paul_at_all-the-johnsons.co.uk)
Date: 01/05/04
- Next message: Ulrich Eckhardt: "Re: initializer is not a constant error ??"
- Previous message: Richard Heathfield: "Re: Do I need any Visual Basic before learning C++?"
- In reply to: Chris \( Val \): "Re: Learning process"
- Next in thread: Chris \( Val \): "Re: Learning process"
- Reply: Chris \( Val \): "Re: Learning process"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Mon, 05 Jan 2004 07:56:51 +0000
Hi,
Chris ( Val ) managed to grunt out the following :
> "Paul F. Johnson" <paul@all-the-johnsons.co.uk> wrote in message
> news:pan.2004.01.04.20.14.44.184479@all-the-johnsons.co.uk...
> | Look at it this way. If you program with the MFC, you can only target
> | Windows users. If you use wxWindows, the same code can be used for ANY
> | platform which has a port of wxWindows, that therefore makes your
> | application *way* more accessible and as I've said increases your
> | potential userbase many many many times.
>
> I don't have any figures at hand, but I totally disagree
> with you - I would bet that MFC actually *dwarfs* the amount
> of 'wxWindows' applications and users out there :-).
Have I said *anywhere* that there are more wxWindows apps out there? What
I've said is that using wxWindows massively increases your userbase
potential.
> | Nope. I've not. What I've said (in effect) is that if he wants to
> | learn something, then learn something which will be of more use than
> | limiting to a single set of libs. Using Qt, GTK+ or wxWindows allows
> | you to produced the moneycow Win32 apps while also allowing your
> | software to other markets - something *anything* propietory doesn't.
>
> Note: I am not endorsing that everyone go out and learn MFC, nor am I
> endorsing that everyone go out and learn wxWindows - it was you who
> brought these topics up in the first place.
Sure did.
> What I am questioning, is your comment, quote:
> "treble your potential client base by using wxWindows"
>
> Sorry, but I find that very difficult to believe - Max :-).
>
> I do not have any statistics, but my feeling is that it is actually the
> opposite.
Okay. Look at Scribus (a Linux DTP app). Uses the Qt libs which means that
it has an instant userbase (for free) of Mac OSX and Linux. There are over
a million users of it (currently, based on downloads). Not a massive
number compared to Quark or Indesign. If it had been designed using
wxWindows, it would have a *potential* userbase of Windows, Mac <OSX, OS
X, Linux, OS/2 and a few other platforms. The same would apply if Qt was
free for Windows. Keyword is potential. Potentially, by using wxWindows,
you're massively increasing the userbase which you 100% won't do using MFC.
> | > Most job adds that I see advertised, *explicitly* state that MFC
> | > knowledge is a requirement(and an advantage), for the position - you
> | > just cannot get away from it.
> |
> | I know :-( Companies are so silly that way. They don't recognise that
> | Windows sales and the Windows userbase is decreasing world wide. But
> | then, how often is it that you see companies move with the times!
>
> Again, it is these kind of quotes that I am against, because you are
> only guessing. Sure, *nix* is increasing it's user base(as one would
> naturally expect), but why do you think the MS user base is declining ?
MS sales figures, number of documented companies moving over and removing
Windows as the desktop environment over companies adopting.
> About 6 months ago, I read in a PC magazine, that yes indeed, Linux did
> make some more sales(and increase its user base), but at the same time,
> MS had again dwarfed these statistics for the same financial period -
> full stop.
I'm not saying Windows is massive, but the tide is turning. However,
that's OT.
> It would take countless years to even get near the 'Redmond Beast', so
> any discouraging talk of a decline is pointless.
Not really - it only takes people to wake up to the "let's patch again,
like we did yesterday" mentality of MS - and people are.
> As much as you or any others despise MS, you do at least have to
> acknowledge that they have made significant improvements, are continuing
> to do so, and *do have an excellent product*.
Nothing MS produce is excellent in anyway, shape or form. If a box comes
preinstalled with an OS, people stick with it - lots never bother applying
patches. It usually doesn't matter how bad the product is either. Face it,
Outlook/Outlook Express are two of the most insecure piles of pap on the
face of the planet - yet will people move to Pegasus or EudoraLite? No.
They don't need to as they already have O/OE. Mozilla is now a far more
capable browser than IE - it is certainly more compliant to w3c specs - is
it dominant? No. Why? It's not bundled with Windows.
A stranglehold does not make for "improvements" either. Win3.11 was the
most stable version of Windows with Win2K being the most usable, yet the
current version has more holes than a cheese grater, tries to do the
thinking for the user and treats everyone like a 3 year old (it's not for
nothing that XP was dubbed the Teletubbie edition in the UK)
> | Depends. The object model for wxWindows is not greatly different to
> | MFC and quite a lot of the methods have an equivalent (or close to
> | one). If you can make a case to the PHB that to transfer over to wx
> | will *make* the company even more money at the cost of probably a
> | couple of weeks development time, he/she would have to be crazy not to
> | say go with it.
>
> On the contrary, they would not be the crazy ones :-).
Given the choice of two weeks work in return for lots of money, or keep
plodding on is crazy? Makes sense to do the work!
> Are you now going to tell me, that you can re-write an application,
> re-document it all(ERs, DFDs etc...), unit test the code, go through
> UAT, etc.. it a couple of weeks ? LOL :-).
Depends. It is not unknown for large programs based on MFC to be moved
over in 3 weeks. Once the test procedure is there (and lots of the same
ones *can* be used), things slot into place.
> I wasn't talking about a toy program you know!
I know. Toy ones tend to take ages!
> | I doubt I'd go looking for another job as to be in that job in the
> | first place I'd have to know MFC ;-p
>
> Ahha! So it isn't pointless knowing MFC or any other technologies after
> all, and you have decreased your employment opportunities.
I never said it was pointless using MFC. That's daft. I've said that it's
pointless learning it if you're never going to use it.
> | > 2. What if you went to an interview, and the interviewer told
> | > you that they use MFC, and are prepared to train(or allow you to
> | > learn it) - Would you seriously not take the position ?
> |
> | Depends on the company and if they actually had a well thought out
> | reason (and not some dogmatic PHB reason thought out by some drone in
> | Redmond) for using MFC.
>
> Depends on what ? - at the interview, they are telling you:
>
> *Hey Paul, MFC is what we use(because we have a lot of code
> already written in it, so it needs to be maintained), are you happy to
> do that with us ?*
>
> Now there' their reason - what do you do ?
Dunno. Be honest is usually the best policy.
> In the case of a newbie(looking for a break), it would be idiotic not to
> take it - you can always learn on the job and look elsewhere whilst
> gaining experience, because there will surely be other things to
> learn(apart from MFC).
In the UK, we have the catch 22 situation. Not many places actually teach
MFC! I don't know any NW institution which teaches MFC for 16 - 19 year
olds or HE places which do it either. I'd even go as far as to say that
most MFC bods have either learned on the job *or* learned on their own
time.
> | True. Over the years, I've learned and used many libraries - it's part
> | of the fun. However, if (in the case of the OP) he/she is asking what
> | to learn next, deciding on a proprietory route may not be the best
> | idea.
>
> Finally, I agree with you - but I never mentioned anything about the OP
> taking that proprietary route, you did initially, and my job was to
> debunk those myths :-).
Fair enough. Nice to see we're agreeing again ;-p
> | > Should we also tell the OP to forget about good old fashioned MS-DOS
> | > ?
> |
> | Yes. DOS really should have been shot at birth
>
> But the windows command line is still widely used today. You *have* to
> know your way around it, even if using *nix*.
Windows has not really had a command line environment since DOS 7 vanished
(XP). It is not that widely used any more.
> Please do not confuse MS-DOS with today's Windows command line.
I'm not. DOS was usable, it had it's problems, but was definately more
usable!
> | > or working with today's modern command interpreters under Windows ?
> |
> | Work under the best thing for you. Look, if someone wants to use
> | MSVC++ that's up to them, if someone wants a truely transferable
> | skill, then MFC isn't what they're after.
>
> Yes, that is true, but you're misinterpreting what I am saying.
>
> I'm saying - learn everything and anything that will make you not only a
> better person, but a much more established programmer or development
> engineer in the long run.
Which is fair enough - by if you're going to be a (in my case, at least
by training) Physical Chemist, why should I learn sociology? The same
applies here. If someone is going to be based in the *nix environment, why
should they bother with MFC?
> The more tools you have in your backpack, the better the chance that you
> will solve any given problem(s), quickly and efficiently, using the
> right selection of tool.
Correct. This is where a lot of folks (IMO) fall down today by the sheer
monoculture out there. You have one set of tools, one inefficient set of
libs and anything outside of that is "big and bad" (paraphrasing a college
lecturer friend)
> | > These are imperative skills in any workplace environment. What do
> | > you tell your boss ?
> | > Er, sorry - I only know how to do it under 'bash' ?
> |
> | Depends where you work ;-p
>
> Bzzzzzzzzzt! - You've just been sacked :-).
Nope. bash (and csh and zsh) are all quite happy together. Anyone using
bash understands that - even PHBs!
> | Same applies to any technology though. In the UK we have loads of
> | people between 25 and 32 who were brought up on RISC OS. If they've
> | learned to program, they will have used the Acorn libraries. But these
> | are fewer and further between than anything. Same applies to those who
> | limited themselves only to traditional MacOS, OS/2 and DOS - their day
> | has been and gone. W32 is suffering now and I have no doubt that in 10
> | years time, Linux and BSD will be suffering in favour of something
> | else....
>
> Yes - It's called 'LogHorn' :-).
Longhorn is awful. Have you seen the betas? Euw!!!!! Longhorn, especially
created for those who can just about turn the computer on and like eye
candy and don't mind everything running dog slow!
> | > My motto - *keep an open mind*, and don't shut doors on any kind of
> | > learning in life - having(and increasing your knowledge) is a good
> | > thing, and hardly pointless at all.
> |
> | It is only pointless if you're never going to use it!
>
> But I told you in my earlier example, I will never become an astronaut,
> nor scientist - is my happiness pointless ?
Your talking of a passtime, not a profession. I enjoy writing music, but
never intend to be a musician.
TTFN
Paul
-- One OS to fool them all One browser to find them One email client to bring them all And through security holes, blind them...
- Next message: Ulrich Eckhardt: "Re: initializer is not a constant error ??"
- Previous message: Richard Heathfield: "Re: Do I need any Visual Basic before learning C++?"
- In reply to: Chris \( Val \): "Re: Learning process"
- Next in thread: Chris \( Val \): "Re: Learning process"
- Reply: Chris \( Val \): "Re: Learning process"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
