11 Dec
2008
11 Dec
'08
11:27 a.m.
Tony Duell wrote:
I've never been on a business trip, and probably never will, so that's
not an issue for me. In any case, 1000 pages is not that thick a pile of
paper.
And if I'm doing serious design, I like to have the device in front of
me, a soldering iron, logic analyser, etc. In other words, I'd want to
investigage the device. Not something that's easy to do away from the bench.
Of course paper has the advantage that I can read it anywhere, it never
crashes, it never gives strange error messages, etc. And it withstands a
hot soldering iron a lot better than any computer does :-)
On the other hand, it's hard to drag 1000 pages of doc on a business
trip because you want to do some hobby coding/design during the boring evenings, or if you do work on the road.
Ypu miounderstnad me. If I want to find a particular device -- by type
number -- then I have no problem typing the number into one of the
approriate web pages and looking at the pdf data sheet. Well, one
problem, which I'll mention later.
But I routinely flip through databooks looking for 'interesting' devices.
Not necessarily from the highlighted specifications. Or even the data
sheet on that device. I might spot an application circuit, see some
device in it that I've not heard of, look it up, and then file away in
the back of my brain that <foo company> makes a useful chip to do <bar>,
the data sheet is in book <baz>. And I'll rememebr that when I have that
sort of problem to solve.
I have yet to find a way to do that efficiently with pdf files.
And I simply can't 'flip through' a
pdf file (even on the fastest
machines I've ever used) in the same way that I can flip through a book.
I used to routinely flip through data books, etc and interesting devices
would catch my eye. It's much more difficult doing that with pdf files.
"flip"? You really are oldskool. If you need to go to a section,
there's search, and a specific page is but a few numbers into the "Page"
box away. -tony
PDF can be our common ground though, You print it out, I read it on
screen. I like formats that allow us both to work effectively.
Can't go with the "TXT FOREVAR" folks, though.
And to your other point, to be consistent, I must concur. It's a hobby,
and anything goes. I was only asking that a disclaimer be applied,
since the difference between "I code this up on my own because it's a
hobby and I think it's fun" and "I don't trust pre-written apps and you
should not either. Write them yourself all times" is sometimes hard to
distinguish.
I didn't comment on your earlier message on this, but I feel you're
misguided. Well, that's putting it politely, I won't use the words I was
thinking of on a public list :-)
I have _never_ found a computer or electronics company that offers what
I consider to be even marginally adequate technical support. I've
contacts many, both privately, when working at well-known universities,
and when working at large, well-known companies. Without exception they
have been useless. That's one reason I insist on being able to support
everything I depend on myself (meaning that I have schematics, source
code, etc).
I feel I (and most, if not all other desingers/programmers) are most
productive if they are using the right tools. Those might be commercial
tools, they might be home-made. Nobody would expect an electronics
designer to make his own 'scope or multimeter before starting out. Noody
would expect a machinist to start by having to make the lathe. But every
electronics designer that I know has a number of home-made devices to
simplify certain jobs, praabluy as add-ons to commerical instruments.
Every machinist makes add-ons for the lathe (In fact I've read plenty of
books on things like clockmaking which point out that many such special
tools are simply not avaiale commercially, you have to make them). When
it comes to programming, I wouldn't expect to have to write a C compiler
(unless I was being asked to develop tools for a new processor :-)). But
there are plenty of samller jobs where it can easily make sense to write
the necessary program from scratch. A trivial example. I had to convert
some ROM dumps from one format to another. There may be a program that
does just what I want if I give it the right options. But by the time
I've found the program, worked out what those options are, etc, I might
as well have writen the 10 lines or so of C to do the job.
However, I do feel that any 'tool' like this needs to be supportable. A
progam needs comments to say what it does, and any odd features. A bit of
hardware needs a scheamtic. Any engineer/programmer who can't support
something as simple as this given comments/a schematic is quite frankly
in the wrong job.
-tony