Skip to main content
news

Re: Paintshop and Corel

Sandman
SubjectRe: Paintshop and Corel
FromSandman
Date11/29/2013 07:35 (11/29/2013 07:35)
Message-ID<slrnl9gdg6.99f.mr@irc.sandman.net>
Client
Newsgroupsrec.photo.digital
FollowsEric Stevens

In article <gpgf99pcs9vdum7gifh0bnjc18ti5pbfk0@4ax.com>, Eric Stevens wrote:

Sandman
That's a very ironic claim, Eric.

The topic is backup software, and Tony's insistance of forcing the word "protocol" in different ways in relation to it, ever changing the meaning and definition of the word and how the user and the developer in relation to it.

I have no problem using the word "protocol" in relation to a backup process, but Tony was very condescending towards nospam who hadn't heard it used that way, so I poked him to see if he could explain further. He couldn't and is now stuck in a death spiral contradicting himself with every post he makes.

Eric Stevens
You mean, you split hairs and twist words for the purpose of further tangling the issue.

No, I don't mean that.

Sandman
A "protocol", in this context, is a procedure in which a given task should be carried out. You have a list of actions to go through in the protocol. This is nothing weird or even strange even when it comes to backup. But it is unusual to refer to an automatic backup function as a "backup protocol" since by the very fact that it's automatic it means there are no steps to take. It is done automatically.

Eric Stevens
You are not thinking far enough back up the chain. A protocol establishes what should be done and possibly how, in broad terms, it should be done.

It can, yes. It can also be quite simpe, or quite shallow.

In the case of the single button backup, the designer's protocol says the user should be given a simple means of backup of a defined list of files (unknown to me) to a defined destination (also unknown to me).

You are free to call it what you will Eric - the *developers* themselves doesn't refer to this as a "protocol". Another laymen moniker for the "black box" you don't understand.

The programmer then gets to work and writes the appropriate code. The protocol is written without any knowledge of the code that is going to be used to implement it.

This is just ludicrous. Many sogftware projects follow a timeline, or flowchart kind of process. Few follow a "protocol". It's not like you have a software designer setting up a "protocol" and the software engineer following that protocol. They usually set up flowcharts, like this one:

http://sandman.net/pages/webblog/2009-09-09/Layoutmotor_For_Atlas

That's a flowchart of relation between functions and actions. You could also set up flowcharts for a conditional software flow (i.e. if/else flowchart)

The user's backup protocol for the single button backup might say 'I shall use the single button backup every day at lunch time' or 'once a week' or 'whenever I have done something important' or 'I'm never going to use it'.

"I'm never going to use it" is not a protocol, Eric. Neither are any of the statement you just said. That's just a schedule. It's hilariously ironic that you were all high and mighty about me not understanding the word "protocol" when you just said that a backup protocol can be to never use a backup function!

Again, a protocol is a list of actions you are to take in order to fulfill a given task. This *usually* refer to a manual scenario.

I.e. starting your car isn't considered a "protocol" because you just turn the key and it does the rest for you using the start engine - but before the time of start engines and other automated parts of modern engines, starting a car most certainly required a protocol - you would saturate the gas lines, put the gear box in neutral and crank the shaft and things like that. It had to be done in a certain order for it to actually work.

The user's single button backup protocol neither requires nor makes any use of the code used to impliment the single button backup.

Wtf? I think you're trying your darnest to make even less sense than Tony.

Sandman
Tony tried to retcon this into sayiing that the protocol is "internal" in the backup function and implemented by the developer, at which point I - a developer (and others) - pointed out that programmers rarely refer to their internal code structure as "protocols", so it was still an unusual word to be used. Not incorrect or even wrong - just unusual - and as I've said, the kind of term a layman would use to describe something he doesn't know anything about.

Eric Stevens
Of course programmers 'rarely refer to their internal code structure as "protocols"' for the simple reason that they aren't protocols. They implement the protocols which, as a result, end up being built into their very structure.

Sigh. No, developers don't "implement the protocols". Whenever a programmer need to follow certain steps to achieve an end result, it's called a conditional process, which is far far more complex than a mere protocol.

Stop trying to shoehorn a word into places and areas you have no knowledge about.

-- Sandman[.net]