When you need to know... what do you do?

[Brad Jones]

When you need to know something technical that you don't already know....
what do you do? Before answering, please think about it for a minute. I'd like to know what you do more so than what you think (or should) you do! Actually, I'd also be interested in knowing what you should do as well.

If you answer, be aware that you may be quoted outside of this forum

A similar question - if a high school student asked you how to research C#, what would you tell them to do? (What if they were new to C#? What if they already knew C#?)

Thanks for your input on this!

Brad!

[darwen]

I've always considered that in order to know what you need to know about a subject you must first learn enough to realise what you don't know.

Wisdom comes not from knowing, or from not knowing but from knowing enough to know what you don't know.

Take a developer's position for instance. Mr Boss comes in one day and says "we need to do... blah blah blah - how long is it going to take you ?"

If this is concerning an area in which the developer has no knowledge of this is a two stage process :

(1) Estimating how long it'll take him to learn enough about the area to be able to give an estimate.
(2) Estimating how long the implementation phase of the project will be.

(1) is obviously practically impossible. However the developer can base their assumptions that the area is similar to other areas which they've had to learn and can base the estimate on this.

Only when (1) is completed can (2) be attempted.

That's how I always base my learning. I use MSDN/Google to bring my level up to a place where I know what I don't know.

Take my learning of C# for instance. The first stage was to gather enough information together to get a 'feel' for the language. Obviously test applications played a huge part in this.

Past experience played a big part too. From knowing MFC very well I knew what was possible and impossible and therefore didn't waste my time trying to find ways in C# to add buttons to notepad and that sort of thing.

Once I'd learned how to get C# to do what I knew MFC was capable of, I was then able to go learn about things which .NET supports but MFC/C++ doesn't e.g. reflection. These had been brought up by my investigations into getting .NET to do the same thing as MFC.

I also find it useful to keep a little notebook of "to be researched later" items when learning things. You'll often come across references (purely by chance) which you haven't the foggiest idea about. Jot them down so you won't forget to come back and investigate later. This is always worthwhile. And this is a part of "knowing what you don't know".

Darwen.

[Yves M]

In my case, I first check whether I know anybody who knows this topic already. If yes, I'll just ask and get into a discussion to find out what kind of resources (e.g. MSDN, tutorials, books etc.) I'll need. If no, then I'll start off by reading the blurb about the technology from the web (if it's from a manufacturer who has set up a documentation page, I'll check that out too) and then it's over to google and other people's tests, testimonials, examples and tutorials.

[NoHero]

When you need to know something technical that you don't already know....
what do you do? Before answering, please think about it for a minute. I'd like to know what you do more so than what you think (or should) you do! Actually, I'd also be interested in knowing what you should do as well.

Well... That depends... I am not used to have someone to ask or even to learn from, so Codeguru was quite a new environment for me. I didn't have to learn nearly everything autodidactic. That was my learning style before I hit this forum here. I also only got internet one year before my CG registration. That means I have learned myself C++ & VB on my own. Just by coding something, debugging, trying, suffering a lot of failures and finding a lot of blind valleys. But every little light of success gave me a kick to go on with my work. And know here I am: Still in learning state. I don't want to say I am a (good) programmer yet, because everyone who claims to be something, stopped to become something.

Nowadays I have my books, this forum here and a lot of buddies who are helping me out. I am still suffering failures and blind valeys, but the time to get out of these decreased by the things I have counted before.

The perfect moment, to say: Thank you. And it comes from the heart...

[kirants]

Hm.. I would first find info on that stuff on the internet.
From there, after some initial overview, I would look at what basic knowledge I need to have for understanding that stuff and then plan to start from there.

For e.g. if I were asked "how would I learn MFC" , I would look at what basic knowldge is required to learn MFC in the first place , before even trying to go into MFC. In this case, it would be , knowledge of C programming, Knowledge of Windows programming, knowledge of OOP , knowledge of C++.

If one isn't comfortable with basics, it becomes all the more harder to understand higher level stuff.

Well, of course there is always the problem of knowing what the basics is, where on can always go deeper and deeper. But, there comes a level were not knowing something wouldn't stop you from understanding something. For e.g. one doesn't need to necessarily know assembly language programming to learn MFC.

[Andy Tacker]

When you need to know something technical that you don't already know....
what do you do? Before answering, please think about it for a minute. I'd like to know what you do more so than what you think (or should) you do! Actually, I'd also be interested in knowing what you should do as well.
Brad!

Four Stages of knowlegde and me:

I know that I know.
I know that I don't know
I don't know that I know.
I don't know that I dont' know.

so, Let's start for this point of view...

Case I is perfect. I know and i do.

Case II is a little tough. Here TIME plays a big role. I generally prefer "HIT & TRY", that is try and try again, till you get where you want. If I am short of time, I would go over net and search the required material.

Case III is a little more tough. but when I run into something that "I didnt know that I know", I would be a little relieved and be proud at times.

Case IV is beyond discussion...

A similar question - if a high school student asked you how to research C#, what would you tell them to do? (What if they were new to C#? What if they already knew C#?)Brad!

In that case, mostly I would prefer people to explore MSDN. I, and I hope many others would agree, MSDN is one of the best resources available OFFLINE. Second is searching a dedicated community like ours [CodeGuru]. Thirdly, I would suggest blind searching using search engines.

If a high school student is coming up to me and asking me what should be done to learn C#. hmm, what would i say? I may not be interested in knowing how much knowledge of programming he's got. or otherwise, he might not be interested to answer my basic programming questions. so, what i wrote above is fully applied in this situation.

In my professional life , I would say that I am a very bad "Pointer". I cannot really point someone to choose a way. so, if you've got a problem, its much easier for me to sit together and solve it, rather than diving directions. and believe me that property of mine is highly rated by my colleagues

Too long a post... I hope I am not talking rubbish

[Brad Jones]

Interesting responses so far. What are some of the common places you look for information? I've heard:

• MSDN
• CodeGuru and other communities
• friends/co-workers
• trial and error

[Vanaj]

1. Trial and Error
2. MSDN
3. See #1
4. Code Guru FAQ
5. See #1
6. Code Guru Forums
7. See #1
8. See #6
9. See #1
10. Sleep on it

[Thread1]

In my situation, my first step is to gather basic information from the internet (ie google). From there I can now estimate if I am able to do the work by myself or need further consultation to some other sites. Since I develop programs with MS products my first hangout is the MSDN especially if it needs specifications. I think a post to codeguru is my last resort if I didn't get it to work with the smoking trial and error.

1. Google
2. MSDN
3. Trial and Error
*4. CodeGuru
*5. Pack up things before leaving the office, and will try to solve it the other day.

[Sahir]

What's "smoking trial and error" ?

[Shuja Ali]

For me when I need to know something.. I go through these steps:
1. Try Yourself
2. Google
3. MSDN(Local) as well as Online
4. CODEGURU (And Other Forums)
5. SOMEONE Who Already knows about what i want to learn

1 --> I like this step more coz, here you tend to learn more than u actually wanted to.. Sometimes becomes Time Consuming though(Bosses get angry sometimes )
2 --> In this step even if i don't get the perfect answer, but still i get a feel of what i shud try to get it working
3 --> Nice place to learn.. Usually my search ends there
4 --> Forums are the place where u will learn things which other 3 steps won't teach u...
5 --> Final solution is to ask someone and get his/her help.. (I Hate this.. coz this way u don't learn much)

Having said that, I think the best way to learn is to try it yourself first and if you are not able to find a solution then try scratching internet and other options mentioned above.

A similar question - if a high school student asked you how to research C#, what would you tell them to do? (What if they were new to C#? What if they already knew C#?)

If he is totally new to C#, i would suggest to get a book which will teach him basics.
If he has worked on C# then MSDN and the Samples are the place where he will learn more..

PS: This is what I have learnt so far.. " If your basics are clear, then nothing is impossible. But if you are basically confused then everything is impossible"

Hope i didn't spoil the thread...

[ovidiucucu]

I hate inflexible schemes, even sometimes I have no choice and must follow them.
Strictly regarding the problem posted here, I still can breathe: the answer is depends on situation.

[Hobson]

When I need to find something out...

1. Documentation shipped with source of my trouble (MSDN for VC++, and so on)
2. Google, getting tuts and samples
3. Use Search option at CodeGuru site
4. Question at CodeGuru forum
5. Forum dedicated to my problem (this point is sometimes is 3, but I often end up in places where people cannot help me or dead forums/mailing lists)
6. Trial and error, until I get bored and try something from 1-5 or go to 7
7. Ask other people (friends, I am not working yet, so I have no co-workers)
8. Hit the pillow (or bottle , depends on day time)
9. goto 1

Generally, I try to gather as much information as possible, especially tutorials and samples. I do not mention books here, because in Poland they are expensive and not too specific, however some of them I found very useful.

Hob

[Thread1]

What's "smoking trial and error" ?

It is something like an intense testing (exaggerate) that dusts (let say in keyboard) almost visible and seem like smoke in the air.