From martin.hepp at ebusiness-unibw.org  Thu Aug  2 16:10:13 2012
From: martin.hepp at ebusiness-unibw.org (Martin Hepp)
Date: Thu, 2 Aug 2012 16:10:13 +0200
Subject: [goodrelations] RDFa-markup
In-Reply-To: <9B3E7C647D4D41A5A7E16643753E32BD@SteviePC>
References: <9B3E7C647D4D41A5A7E16643753E32BD@SteviePC>
Message-ID: <327F611D-3FBD-4D34-B462-DF092A7084F1@ebusiness-unibw.org>

Dear Steve:

On Jul 30, 2012, at 1:35 PM, Lists wrote:

> Hello gr community,
> 
> is there a workaround for "gr:eligibleRegions" available?
> I mean there is a bunch of RDFa markup which makes my site slow down- If I
> want to tell the clients all eligble regions.
> Since my products are available in nearly every country of the world through
> the web, gr:eligibleRegions content="ALL" would be helpfull.
> Are there any aproaches available? Do I have to markup every country/region
> of the world?


First of all, it is a misconception that the current pattern makes your page *slow*. Even if you list 200+ countries, using 200 lines of markup

<div property="gr:eligibleRegions" content="DE"></div>
<div property="gr:eligibleRegions" content="DK"></div>
<div property="gr:eligibleRegions" content="IT"></div>
<div property="gr:eligibleRegions" content="ES"></div>
...

this does not make your page slow, even if it looks verbose.

The reason is that HTTP 1.1 natively supports compression, which means that the serving server will "zip" the markup before sending it over the wire and the client will unzip it on the fly. Since the lines are so similar, the compression rate will be in the order of 90 % or more.

So from the payload that is being sent over the wire, the actual markup will not require significantly more bytes than a syntax that would not repeat the property multiple times.

Second, most pages have a total size of 300k to 1 MB, if you properly count all Javascript and images, while even comprehensive RDFa markup hardly goes beyond 5 - 10 KB extra content. Plus the RDFa compresses much better than Javascript and images. 

So there is no reason to worry. 

If you want to speed up  your pages, 
- setting the cache parameters wisely, 
- tuning database queries and index structures of your database (e.g. mySQL), 
- reducing the amount of Javascript inclusions (both in terms of the number of files and their sizes), and 
- reducing the number of CSS files are much more effective and relevant measures.

See also [1].


As a side note: From a SEO perpective, it does not make too much sense to claim that you deliver to all countries in the world but restrict your list to those to which you actually do business to, because you want to send specific relevance signals to the search engines. So instead of a plain list of all countries, use those 10 - 50 that you actually consider target regions.

I hope this helps.

Best wishes

Martin Hepp


PS: In the next service update of GoodRelations, you will be able to use http://schema.org/GeoShape for indicating the eligible shipping regions. So you could define the Earth's surface as such and use that.

[1] http://wiki.goodrelations-vocabulary.org/Frequently_Asked_Questions#Will_the_extra_markup_slow_down_my_page.3F


On Jul 30, 2012, at 1:35 PM, Lists wrote:

> Hello gr community,
> 
> is there a workaround for "gr:eligibleRegions" available?
> I mean there is a bunch of RDFa markup which makes my site slow down- If I
> want to tell the clients all eligble regions.
> Since my products are available in nearly every country of the world through
> the web, gr:eligibleRegions content="ALL" would be helpfull.
> Are there any aproaches available? Do I have to markup every country/region
> of the world?
> 
> c u steve
> 
> 
> _______________________________________________
> goodrelations mailing list
> goodrelations at ebusiness-unibw.org
> http://ebusiness-unibw.org/cgi-bin/mailman/listinfo/goodrelations



--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp at ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
=================================================================
* Project Main Page: http://purl.org/goodrelations/






From lists at psycholutions.com  Fri Aug  3 03:40:12 2012
From: lists at psycholutions.com (Lists)
Date: Fri, 3 Aug 2012 03:40:12 +0200
Subject: [goodrelations] RDFa-markup
Message-ID: <3695BEE93C4D4B4BA20F12E05B0E2026@SteviePC>

Thank you Mr. Hepp for answering.

Perhaps I missunderstood the meaning of the item "eligibleRegions". I' am
working on a semantic shop application and will 
sell software with it. So what would search engines do with the informations
for "eligibleRegions" in context of a software product offer?
What is the actual meaning of the item in this context?

I also want to try to get the complete benefit of gr- more then tagging. Are
there any good resources (tutorials, real world examples, videos etc.)
available to build up or use ontologies based on GoodRelations? I found
http://code.google.com/p/gr4php/ and its very interesting. But I need more
"real world input" (ontology usage examples) to get the complete impression
and maybe new ideas :-). 

Will high-precision identifiers for product types based on Wikipedia benefit
from Wikidata?

thx in advanced Steve




From martin.hepp at ebusiness-unibw.org  Fri Aug  3 04:57:32 2012
From: martin.hepp at ebusiness-unibw.org (Martin Hepp)
Date: Fri, 3 Aug 2012 04:57:32 +0200
Subject: [goodrelations] RDFa-markup
In-Reply-To: <3695BEE93C4D4B4BA20F12E05B0E2026@SteviePC>
References: <3695BEE93C4D4B4BA20F12E05B0E2026@SteviePC>
Message-ID: <5C34BD56-C3C2-4214-9F55-BE880950099D@ebusiness-unibw.org>

Hi,
On Aug 3, 2012, at 3:40 AM, Lists wrote:

> Thank you Mr. Hepp for answering.
> 
> Perhaps I missunderstood the meaning of the item "eligibleRegions".
The property is for indicating the regions that you are selling to. 

> I' am
> working on a semantic shop application and will 
> sell software with it. So what would search engines do with the informations
> for "eligibleRegions" in context of a software product offer?
Im general, search engines use GoodRelations markup for three purposes:

1. Direct information extraction, i.e. getting data for specific purposes (e.g. showing Rich Snippets),
2. Getting relevance signals from the site (e.g. for which queries from which audiences should this page be ranked high)?), and
3. for training other components, in particular machine learning functionality.
> What is the actual meaning of the item in this context?
See above.
> 
> I also want to try to get the complete benefit of gr- more then tagging. Are
> there any good resources (tutorials, real world examples, videos etc.)
Tutorials:
    http://wiki.goodrelations-vocabulary.org/Events
    http://wiki.goodrelations-vocabulary.org/Documentation (work in progress)

Examples:
    http://wiki.goodrelations-vocabulary.org/Cookbook
    http://www.heppnetz.de/ontologies/goodrelations/v1.html
    http://wiki.goodrelations-vocabulary.org/Datasets

Videos:
    https://vimeo.com/user2627141

> available to build up or use ontologies based on GoodRelations? I found
> http://code.google.com/p/gr4php/ and its very interesting. But I need more
> "real world input" (ontology usage examples) to get the complete impression
> and maybe new ideas :-). 
> 
> Will high-precision identifiers for product types based on Wikipedia benefit
> from Wikidata?
> 
Maybe. In some cases, you could directly use Wikidata URIs for the same purpose.

Best wishes

Martin Hepp

> thx in advanced Steve
> 
> 
> _______________________________________________
> goodrelations mailing list
> goodrelations at ebusiness-unibw.org
> http://ebusiness-unibw.org/cgi-bin/mailman/listinfo/goodrelations









From martin.hepp at ebusiness-unibw.org  Tue Aug  7 11:49:15 2012
From: martin.hepp at ebusiness-unibw.org (Martin Hepp)
Date: Tue, 7 Aug 2012 11:49:15 +0200
Subject: [goodrelations] New GoodRelations Recipe for Classified Ads
Message-ID: <DEE4F2B5-A536-4C42-979C-5D5C218C714E@ebusiness-unibw.org>

Dear all:

I just added a recipe for classified ads and similar site types to the GoodRelations cookbook:

http://wiki.goodrelations-vocabulary.org/Cookbook/Classifieds

Any feedback is very welcome!

Best wishes

Martin Hepp

--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp at ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
=================================================================
* Project Main Page: http://purl.org/goodrelations/






From martin.hepp at ebusiness-unibw.org  Wed Aug 15 10:34:03 2012
From: martin.hepp at ebusiness-unibw.org (Martin Hepp)
Date: Wed, 15 Aug 2012 10:34:03 +0200
Subject: [goodrelations] GoodRelations vs. SPARQL endpoints
References: <3287C9E6-564D-479F-9946-28B79C7C2321@gmail.com>
Message-ID: <D80819B0-0388-4008-95E0-5101137A6497@ebusiness-unibw.org>

Enclosed, please find my reply to a bunch of questions related to SPARQL endpoints sent to me as private communication. I thought the answers might me relevant for the readers of this list, too:


Begin forwarded message:

> Dear XYZ,
> Thanks for your email. I hope to answer your questions as follows:
> 
> 1. The GoodRelations ontology is just the **schema** for e-commerce data and that is what is available from http://purl.org/goodrelations/v1.owl. Product, company, and offer data published by using the terms in GoodRelations is spread all over the Web. For a list of examples (yet terribly incomplete), see
> 
>    http://wiki.goodrelations-vocabulary.org/Datasets
> 
> Note that ontologies are NOT knowledge-bases, so by definition, they contain no instance data (unless the individuals are part of the schema, like a day of the week). For background information, see
> 
> Hepp, Martin: Ontologies: State of the Art, Business Potential, and Grand Challenges, in: Hepp, M.; De Leenheer, P.; de Moor, A.; Sure,Y. (Eds.): Ontology Management: Semantic Web, Semantic Web Services, and Business Applications, ISBN 978-0-387-69899-1, Springer, 2007, pp. 3-22.
> PDF from http://www.heppnetz.de/files/hepp-ontologies-state-of-the%20art.pdf
> 
> 2. We do not operate a SPARQL endpoint. There are various public SPARQL endpoints that contain some GoodRelations data, like e.g.
> 	? http://sparql.sindice.com/
> 	? http://uriburner.com/sparql
> 	? http://lod.openlinksw.com/sparql
> 	? http://loc.openlinksw.com/sparql
> 
> 
> However, there is currently no "Google-like" RDF repository that offers you free access to the whole Web via a convenient SPARQL endpoint.
> 
> If the repositories from above do not meet the needs for your application, you will have to crawl and cleanse the data yourself, e.g. from
> 	? http://wiki.goodrelations-vocabulary.org/Datasets
> 
> Note that crawling crawling, cleansing, and consolidating structured data at Web scale is a non-trivial task, so do not expect someone else to do it for you for free. It may be a good business idea to offer such a service.
> None of the currently known free and public SPARQL endpoints provide a sufficient coverage and currency of e-commerce data to power serious e-commerce applications.
> You will have to do that on your own if your business idea requires that.
> 
> See also
> 
> http://wiki.goodrelations-vocabulary.org/Frequently_Asked_Questions#Where_can_I_find_a_free_SPARQL_endpoint_with_all_GoodRelations_data_from_the_Web.3F
> 
> 
> Best wishes
> 
> Martin Hepp
> 

--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp at ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
=================================================================
* Project Main Page: http://purl.org/goodrelations/






From martin.hepp at ebusiness-unibw.org  Wed Aug 15 21:32:19 2012
From: martin.hepp at ebusiness-unibw.org (Martin Hepp)
Date: Wed, 15 Aug 2012 21:32:19 +0200
Subject: [goodrelations] New Book Example for GoodRelations and schema.org
Message-ID: <8BAC1AE3-21B8-47C0-93AA-E3E8757EE15C@ebusiness-unibw.org>

Dear all:

I just created a new recipe in the GoodRelations Cookbook for book offers and book master data:

  http://wiki.goodrelations-vocabulary.org/Cookbook/Books

Any feedback will be greatly appreciated!

Best wishes

Martin Hepp

--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp at ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
        http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
=================================================================
* Project Main Page: http://purl.org/goodrelations/