Re: [uf-discuss] Size considerations (or how to choose)

[Ryan King]

On Oct 21, 2006, at 9:58 AM, Charles Roper wrote:

Can anyone give any real *disadvantages* to using output compression?


If you do it right, none. Some browsers, like IE 5 and maybe 6, have  
problems with compressed, cached JavaScript and other weird edge  
cases. However, most HTTP servers come configured to deal with this  
correctly using content negotiation.


Sure, not everyone runs HTTP server software which implements these  
facilities, but I posit that the 80% case is covered.


-ryan
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Ryan King]

On Oct 21, 2006, at 9:37 AM, Charles Roper wrote:

On 21/10/06, Andy Mabbett <[EMAIL PROTECTED]> wrote:

I don't think the intention is that the raw markup of a uF be
"human-readable first".


This is a good point; what, exactly, should be "human readable  
first?" I always assumed it was the rendered output of the HTML,  
rather than the markup itself. Is this correct?


Actually, the markup is intended to be human readable.

-ryan
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose)

[Ciaran McNulty]

On 10/21/06, Charles Roper <[EMAIL PROTECTED]> wrote:

Can anyone give any real *disadvantages* to using output compression?


The choice to use compression is one of bandwidth vs. processing time.
I have personally had a bad experience with a cut-rate ISP who had
some sort of CPU-usage throttling that meant we had to turn off output
compression on a largeish site.

-Ciaran McNulty
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Andy Mabbett]

In message <[EMAIL PROTECTED]>,
Benjamin West <[EMAIL PROTECTED]> writes

>>From :
>"geo is a 1:1 representation of the "geo" property in the vCard
>standard (RFC2426 (http://www.ietf.org/rfc/rfc2426.txt)) in XHTML"
>
>As you can see, the authors of the spec weren't the ones doing any
>abbreviating.  The name was picked to reuse a pre-existing standard.

I've never suggested otherwise.

>* Are there any examples on the web where people are using "sci" as a
>class name in a way that roughly means what you also intend it to
>mean?

Yes.

However, if you care to look back over the archives and the 'wiki',
you'll see that I've already acknowledged that "sci" should not be used,
and that "taxon" or "biota" would be better; in fact I did so a month
ago tomorrow.


-- 
Andy Mabbett
Say "NO!" to compulsory ID Cards:  

Free Our Data:  
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Benjamin West]

I think this has been mentioned before, but I'll mention it again.


From :

"geo is a 1:1 representation of the "geo" property in the vCard
standard (RFC2426 (http://www.ietf.org/rfc/rfc2426.txt)) in XHTML"

As you can see, the authors of the spec weren't the ones doing any
abbreviating.  The name was picked to reuse a pre-existing standard.

In picking out class names, you might find it fruitful to look at
names already be used to describe the kind of data you hope to mark
up.  This is a core tenant of microformats.

There are some simple tests to resolve debate around whether or not to use sci:
* Are there any examples on the web where people are using "sci" as a
class name in a way that roughly means what you also intend it to
mean?
* Are there any standardized, or conventionalized, formats that
describe what you mean to describe?  Do they use "sci" or something
else?

One effects of networked systems is that re-use raises the efficacy of
both the original and the copy.

Ben
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Andy Mabbett]

In message
<[EMAIL PROTECTED]>, Charles
Roper <[EMAIL PROTECTED]> writes

>> >I think comparing geo and sci, etc. is not a great example as I think
>> >geo can be thought of as a well known abbreviation.
>>
>> Yes, it clearly identifies rocks, to geologists ;-)
>>
>> But seriously, do you really think it's well known, outside of the "web
>> 2.0" community?
>
>I might poll some people round the office (where most people's
>primiary interests biology, conservation and ecology) and see what
>they think. I'm going to make a guess that they'll say "geography" or
>"geographical".

I hope you don't conduct all your experiments with such built in bias
;-) What about asking some non-scientists? And some people called
George!

A Google search for "geo" has just returned, in the top five results,
none with the meaning "geograph*"

An eBay search finds "Geo V" coins and stamps and "Geo III" silver.

Wikipedia  also has a number of other
non-geographical uses.

-- 
Andy Mabbett
Say "NO!" to compulsory ID Cards:  

Free Our Data:  
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose)

[Andy Mabbett]

In message
<[EMAIL PROTECTED]>, Charles
Roper <[EMAIL PROTECTED]> writes

>Can anyone give any real *disadvantages* to using output compression?

Perhaps not - but is it always available to people? Not everyone manages
(or has access to the management of) the servers on which their content
resides.

Nor is file size just a transmission issue - there's also storage, and
the possibility of needing to put files onto media such as memory
sticks, CDs or even floppy discs.

-- 
Andy Mabbett
Say "NO!" to compulsory ID Cards:  

Free Our Data:  
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose)

[Charles Roper]

On 21/10/06, Kevin Marks <[EMAIL PROTECTED]> wrote:

On the broader point, assuming you use gzip when you care about size,
abbreviations don't save much, especially in the many-repeated case
discussed.


This is one of my primary arguments against using abbreviations. See
my original post:

http://snipurl.com/10156

There's an online tool for doing before and after comparisons. Here
are the results from a medium sized page from one of my sites using
hCard:

http://snipurl.com/1015m

That's 26kb down to 5kb - an 82% compression rate and 5.3x transfer
rate improvement.

The compression check page can be found here:

http://www.port80software.com/tools/compresscheck.asp

Can anyone give any real *disadvantages* to using output compression?

Charles
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Charles Roper]

On 21/10/06, Andy Mabbett <[EMAIL PROTECTED]> wrote:

>> And yet we have "geo".
>
>I think comparing geo and sci, etc. is not a great example as I think
>geo can be thought of as a well known abbreviation.

Yes, it clearly identifies rocks, to geologists ;-)

But seriously, do you really think it's well known, outside of the "web
2.0" community?


I might poll some people round the office (where most people's
primiary interests biology, conservation and ecology) and see what
they think. I'm going to make a guess that they'll say "geography" or
"geographical".


>As microformats are human-readable first I think size is a secondary
>consideration.

I don't think the intention is that the raw markup of a uF be
"human-readable first".


This is a good point; what, exactly, should be "human readable first?"
I always assumed it was the rendered output of the HTML, rather than
the markup itself. Is this correct?

Charles
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Andy Mabbett]

In message <[EMAIL PROTECTED]>, Christopher Rines
<[EMAIL PROTECTED]> writes

>> And yet we have "geo".
>
>I think comparing geo and sci, etc. is not a great example as I think
>geo can be thought of as a well known abbreviation.

Yes, it clearly identifies rocks, to geologists ;-)

But seriously, do you really think it's well known, outside of the "web
2.0" community?

>As microformats are human-readable first I think size is a secondary
>consideration.

I don't think the intention is that the raw markup of a uF be
"human-readable first".

-- 
Andy Mabbett
Say "NO!" to compulsory ID Cards:  

Free Our Data:  
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose)

[Kevin Marks]

On Oct 19, 2006, at 9:57 PM, Christopher Rines wrote:


In my opinion amount is a really difficult one to abbreviate (or any 
measure
for that matter) as it can be used to describe a lot of other things 
for
which there is not yet a microformat but cur (for currency) is 
interesting
as just off the top of my head I don't think currency is used in a lot 
of

other situations but could we abbreviate current (if we did something
electrical) with cur?

I guess this reinforces my point that while useful abbreviations in
human-readable things are tricky at best. Just like acronyms can be an
insiders language, abbreviations can obfuscate meaning.


On the broader point, assuming you use gzip when you care about size, 
abbreviations don't save much, especially in the many-repeated case 
discussed.


Also, using whole english words means that the gzip is likely to be 
more able to exploit the natural redundancy of the English language 
(assuming your text outside the markup is English).


If someone has whole-page examples I'll be happy to compare before and 
after gzipped ones.


___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: RE: [uf-discuss] Size considerations (or how to choose abbreviations)

[Charles Roper]

On 20/10/06, Brian Suda <[EMAIL PROTECTED]> wrote:

--- the tricky thing is that there are no namespaces in Microformats,
so if you use cur, sure it is scopped to 'money', but it is now a
'reserved word' for all of microformats. As it was pointed out in a
previous message, then what happens to 'cursor' or 'current', you
can't do: now()
even thought that 'cur' is scoped to some ficticious 'time', it still
would mean 'currency'.


This is what I suspected and is why I'm uncomfortable with using "sci"
in the species microformat. I mean, for me, if that were to be used as
an abbreviation, it would be for the word "science" (as in sci-fi),
rather than "scientific name". If the species uF were to grab it, then
no other future format could then use it, even if said future format
needed an abbreviation of the word "sci". All hypothetical, of course,
but you catch my drift.

As microformats proliferate, I think the namespace issue could become
more troublesome. It's best to think carefully about these issues now,
rather than a few months/years down the line when it becomes more of a
problem. I'm personally holding out for "cur" to be used in a curry
microformat. ;)

Cheers,

Charles
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: RE: [uf-discuss] Size considerations (or how to choose abbreviations)

[Brian Suda]

On 10/20/06, Mike Schinkel <[EMAIL PROTECTED]> wrote:

However, and this is an honest question, isn't "currency" and "amount"
really only valid in context with "money?"  Wouldn't that make it okay to
abbreviate the children of money, like so?:


   $
   5.99



--- the tricky thing is that there are no namespaces in Microformats,
so if you use cur, sure it is scopped to 'money', but it is now a
'reserved word' for all of microformats. As it was pointed out in a
previous message, then what happens to 'cursor' or 'current', you
can't do: now()
even thought that 'cur' is scoped to some ficticious 'time', it still
would mean 'currency'.

It is best to be as explicit as possible and use the full names.
Infact, in the world or reuse, that probably should be:


   $
   5.99


ADR, TEL, and others (probably UID) use that structure. It works, no
need to re-invent the wheel and add more terms that mean the same
thing.

-brian

--
brian suda
http://suda.co.uk
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

RE: [uf-discuss] Size considerations (or how to choose)

[Christopher Rines]

Hey Mike,

This is an very good/interesting example...  

In my opinion amount is a really difficult one to abbreviate (or any measure
for that matter) as it can be used to describe a lot of other things for
which there is not yet a microformat but cur (for currency) is interesting
as just off the top of my head I don't think currency is used in a lot of
other situations but could we abbreviate current (if we did something
electrical) with cur?  

I guess this reinforces my point that while useful abbreviations in
human-readable things are tricky at best. Just like acronyms can be an
insiders language, abbreviations can obfuscate meaning.  

To reiterate my position I love abbreviations but anywhere they are used
really need to be studied. :-)

Cheers,

Christopher

In message <[EMAIL PROTECTED]> Mike Schinkel
<[EMAIL PROTECTED]> in addition to other things said:

> I have a question about that.  I'll use the example of money because it's
one I'm more familiar > with.  In this particular case, we have money,
currency, and amount:

>   
>  $
>  5.99
>   
 
> However, and this is an honest question, isn't "currency" and "amount"
> really only valid in context with "money?"  Wouldn't that make it okay to
abbreviate the 
> children of money, like so?:

>   
>  $
>  5.99
>   
 



___
$0 Web Hosting with up to 200MB web space, 1000 MB Transfer
10 Personalized POP and Web E-mail Accounts, and much more.
Signup at www.doteasy.com

___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

RE: [uf-discuss] Size considerations (or how to choose abbreviations)

[Mike Schinkel]

>> The point I am trying to make is abbreviations can be very dangerous and
are very easy to mis-interpret so I think we need to think long and hard
before choosing and implementing them.  I am not arguing against them in
specific cases but very well thought out cases.  

I have a question about that.  I'll use the example of money because it's
one I'm more familiar with.  In this particular case, we have money,
currency, and amount:


   $
   5.99

 
However, and this is an honest question, isn't "currency" and "amount"
really only valid in context with "money?"  Wouldn't that make it okay to
abbreviate the children of money, like so?:


   $
   5.99

 

-Mike

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Christopher Rines
Sent: Thursday, October 19, 2006 10:45 PM
To: microformats-discuss@microformats.org
Subject: Re: [uf-discuss] Size considerations (or how to choose
abbreviations)


In message <[EMAIL PROTECTED]> Charles Roper
<[EMAIL PROTECTED]> in addition to other things said:
> Should "bin", var", "cult", etc., be written in full? (I think not, to 
> save bloating file sizes)

> These abbreviations are absolutely fine within the very narrow domain 
> of biological nomenclature and taxonomy, but expanded out into the 
> wider domain, then they become horribly generic and lose their 
> meaning. Same with using "sci".

In message <[EMAIL PROTECTED]> Andy Mabbett
<[EMAIL PROTECTED]> in addition to other things said: 

> And yet we have "geo".

I think comparing geo and sci, etc. is not a great example as I think geo
can be thought of as a well known abbreviation.  

As with much other microformat work a well known standard or abbreviation
like vcard I think geo can is a (or close to) "standard" so it is a "safe"
abbreviation which I think is what we should be aiming for when creating an
abbreviation of any type.  I do realize GEO is being used by others such as
the Gene Expression Omnibus (GEO) but I THINK what I say holds as geo being
an implied abbreviation standard.

The point I am trying to make is abbreviations can be very dangerous and are
very easy to mis-interpret so I think we need to think long and hard before
choosing and implementing them.  I am not arguing against them in specific
cases but very well thought out cases.  

As microformats are human-readable first I think size is a secondary
consideration.  Are there any stats about how many sites are compression
enabled vs. not?

Thank You,

Christopher



___
$0 Web Hosting with up to 200MB web space, 1000 MB Transfer 10 Personalized
POP and Web E-mail Accounts, and much more.
Signup at www.doteasy.com

___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Re: [uf-discuss] Size considerations (or how to choose abbreviations)

[Christopher Rines]

In message <[EMAIL PROTECTED]> Charles Roper
<[EMAIL PROTECTED]> in addition to other things said:
> Should "bin", var", "cult", etc., be written in full? (I think not, to 
> save bloating file sizes)

> These abbreviations are absolutely fine within the very narrow domain of 
> biological nomenclature and taxonomy, but expanded out into the wider 
> domain, then they become horribly generic and lose their meaning. Same 
> with using "sci".

In message <[EMAIL PROTECTED]> Andy Mabbett
<[EMAIL PROTECTED]> in addition to other things said: 

> And yet we have "geo".

I think comparing geo and sci, etc. is not a great example as I think geo
can be thought of as a well known abbreviation.  

As with much other microformat work a well known standard or abbreviation
like vcard I think geo can is a (or close to) "standard" so it is a "safe"
abbreviation which I think is what we should be aiming for when creating an
abbreviation of any type.  I do realize GEO is being used by others such as
the Gene Expression Omnibus (GEO) but I THINK what I say holds as geo being
an implied abbreviation standard.

The point I am trying to make is abbreviations can be very dangerous and are
very easy to mis-interpret so I think we need to think long and hard before
choosing and implementing them.  I am not arguing against them in specific
cases but very well thought out cases.  

As microformats are human-readable first I think size is a secondary
consideration.  Are there any stats about how many sites are compression
enabled vs. not?

Thank You,

Christopher



___
$0 Web Hosting with up to 200MB web space, 1000 MB Transfer
10 Personalized POP and Web E-mail Accounts, and much more.
Signup at www.doteasy.com

___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss