[techtalk] Re: [issues] Standards?? PHOOEY!!!

[Jenn V.]

JoAnne Abbott wrote:
> Jenn V wrote:

> > Could you define the difference between a protocol and a standard?
> 
> A protocol is a generalised consensus of the way something is done.
> 
> A standard is so many meters, liters or lines of code.

Sorry, I disagree with these definitions.

A (computing) protocol - to me - is a description of how My Program 
communicates with Your Program.

A (computing) standard - to me - is a definition of a protocol, language
(or anything else), which is openly available and enables people to write
(or build) stuff which matches up with other peoples' stuff.

IE: IP is a protocol. AOL's message chatter stuff is a protocol. IRC is a 
protocol.

BUT: IP has a standard. IRC has a standard. I can write stuff and be 
sure that my IP or IRC stuff works with anyone else's. (So long as we're
both compliant to the standard).
AOL's message chatter stuff _doesn't_ have a standard. I can write stuff
for AOL's message system, but I'm only guessing and it might not work.


Now, I am NOT saying 'I'm right, you're wrong'. Repeat: I'm NOT saying 
that.
I AM saying 'we seem to have different definitions, and be talking about
different things'.

This explains why I don't understand your main point. Hmmm.
Keeping my definitions in mind, can you try again to explain? You may
want to read the rest of my responses first.


Note that in the course of this email, I'll use MY definitions. Because
I'm afraid I still don't understand yours.



> > If this bothers you, go right ahead and make a new internet - or 
> > whatever it is that you want. Of course, if you don't interact with 
> > the current protocols and standards, you'll wind up in the same 
> > situation that MSN did, before it
> > accepted that The People preferred choice.
> >
> > Unless you do something truly useful, as well as innovative.
> 
> AAAHHH  a rephrasing of what I said.

Well at least you know someone's listening to you. 

Next question: I'm not in the least bit upset about this fact.
Why does this bother you?


> > Yes, innovation requires experimentation. Yes, experimentation can be
> > inhibited
> > by standards. Standards committees are slow and careful because their main
> > function is reliability, not innovation.
> 
> another rephrasing. I agree!
> 
> > But absolutely _noone_ is stopping you from innovating on your own
> machine,
> 
> Don't know where this came from.

Because I /think/ you're trying to say 'standards committees' emphasis on 
reliability discourages innovation' and I tried to respond 'no it doesn't,
you can experiment and futz around and prove your whatsit on your own 
machines'.
 
Unless you have some definition of 'innovation' that includes 'changing the
standards'.


> > The committee will ask you to prove its usefulness,  help them define the
> > adjustments
> > to the standard, and other such things. But that's the cost of taking
> > innovation
> > into the marketplace.
> 
> After 40 years in the market place I no longer have the desire for
> flagellation.

Where does /this/ come from?


> I guess my point was missed in the main part.
> The constant clamor for more standards 

What constant clamor?

> and the finger pointing when a
> standard is violated 

Of course. How the futz am I supposed to write an X which works 
with Y when their implementations break with the standard?

And even if they publish their 'new standard', how am I supposed
to write one which works with both Y (on the 'new standard') and
Z (on the 'old standard')?

> seems to me to be more of a noise of someone is
> doing something that makes my work harder. 

What is your work?

> All without stating why this
> is bad except that it violates a standard.

Well look at my paragraphs above. Violating standards makes
programmer's work harder.

> To summarise my original post, which I thought the examples would be
> self explanatory.

Sorry, they weren't.

> Well based standards are good for a less chaotic and more profitable
> environment.

Agreed.

> Even the best standards (inhibit, straitjacket, prevent) (choose one or add
> yours)
> innovation.

Disagree.
 
> Standards are for the past or present. Innovation is for the future.

Agree. And standards committees exist to look at innovations, poke at 
them, and attempt to fit the useful and reliable ones in. They're not
perfect, but - IMO - it's better than either no standards, or unmaintained
standards.


In a later email, Joanne wrote:

> If I don't adhere to the standards of C, C++ no one can use my code.

Isn't that precisely _WHY_ violating standards is bad? 

> Yes
> In the LAN marketplace the winner has yet to be decided but is almost
> certain.
>     But if I don't use your LAN why should I use your protocol?

You shouldn't. Unless you want to talk to my LAN. At which point,
you and I had both better be talking a matching standardised protocol.

> The standards are preventing me from making a 128 bit direct connection
> to a video chip as a innovation. I don't have enough room on the worktable
> for Four motherboards.

No they aren't. The  lack of space on your worktable is inhibiting 
it, but the standards aren't preventing it.

The standards _ARE_ preventing you from making a 128 bit direct 
connection to a video chip _and calling it compliant with motherboards
which don't support it_. 

But if you made the innovation, proved it reliable, and had at 
least one motherboard manufacturer agree it was better, you
and the manufacturer could set a new standard. 
But that's primarily a marketplace issue, not a standards issue.

> Innovation is not orderly. It is messy, confusing and upsetting to adjust
> to.
> It is NEW.

Yup. Standards exist to enable Joe Average to actually buy and use
computers. If my parents and brother - laymen, computer-wise - has
to try to live in the computing-world that I live in, they'd 
freak and run a mile and probably never use computers again.

I live in the industry, watching experimentation happen, letting
things shift and flow around me.
They live in the marketplace, and want to be able to buy a computer
today and have it be good enough for their needs for at least five
years.

I don't mind innovations, for _me_. But because I love my brother,
I want there to be standards.

Remember, I'm using MY definition of the word 'standard'.



Jenn V.
-- 
  "We're repairing the coolant loop of a nuclear fusion reactor. 
   This is women's work!"
		Helix, Freefall. http://www.purrsia.com/freefall/

Jenn Vesperman    jenn at simegen.com     http://www.simegen.com/~jenn