Site level DOI or Journal-specific DOI?

[Ojser]

Hi all,

Just wondering about the assignment of DOI, could we use the same DOI for all journals in OJS installation or it should be unique prefix for each journal?

Thank you

[springday]

Hi Ojser,

you're asking about the DOI prefix, right? There's no problem using the same prefix across different journals, but the DOIs themselves certainly should be unique on a article basis. You would normally get the prefix assigned to you by a DOI registration agency. See here for more info: http://www.doi.org/registration_agencies.html

Best wishes,
Kai

[Ojser]

Thanks Kai,
Yes, I mean the prefix.
They claim to be non-profit and the price is:
$275 if you are "poor" (your revenue is less than $1 million)!
Note, this is annual-based price, not paid once only!
Hihii ! I thought it was totally free, like the ISSN number!
If your revenue is as high as $999 999 or as low as $999 or $0 you should pay $275, though!
Their statistic interval is down apparently not to realize the huge difference between the sum...!
What price we have to pay for DOI if it was a profit entreprise?!!
Well, just publish your articles with vol, year and make it clickable, you will make the best DOI, home made!!
Publishing process have been existed for hundreds of years without DOI, so...
Made home DOI is the best...

[springday]

Well, personally I find most DOI services also a bit over-priced, yet I can understand that it's not for free. The system has a world wide lookup-service for localizing the documents that are registered with a DOI in their system. That certainly requires a strong enough server infrastructure that somebody has to pay after all...

Cheers,
Kai

[Ojser]

Well, for me the best and simple way ever to identify articles is:
Name, article title, journal name and date ! (the order may be customized or globally standardized), and this will give the best ever identifier for the web!

Examples:
Kai Springday, "Site level DOI or Journal-sepecifi DOI" , OJS forum, September 12, 2012
OJSer OJSer, "Site level DOI or Journal-sepecifi DOI" , OJS forum, September 12, 2012

It's simple, short, informative and easy to produce and to search and to find!
We can't have the same name, with the same article title published in the same journal at the same date! So this is a unique identifier.

Why do we need volumes, issues, sub-issues...etc., with the extreme ease of a simple click research, now?
Volumes, issues and other publication identifiers... have been created to facilitate the manual research among huge archives in big libraries and shelves, but now with one click you get what you search for!
Of course, Vol, issue, ...etc. are still valid for traditional publishing methods but for the web...I don't see what that method could offer..
I think, DOI is also created for money; maybe it suits other material objects more than abstract web publishing!
By definition DOI stands for "object" and an article is not an object!
We can identify cars, PC, Houses, phones ...by numbers as they are materials but articles published on the web, this doesn't make much sense; we never talk about DOI saying Hey Steve, did you read the article DOI 10.12345/blablabla...published in... on ...! Rather, we often say the article of Steve published in... on .. (date and/or journal name or both).
This is my point of view, anyway.

[JasonNugent]

Well, for me the best and simple way ever to identify articles is:
Name, article title, journal name and date ! (the order may be customized or globally standardized), and this will give the best ever identifier for the web!

The whole premise behind DOI is that you have 'durable' links that do not change. The items that you've mentioned above can change. They are also not unique (there may be more than one John Smith in the world who has written a paper called 'Widgets', for example). Technically, not even traditional URLs can be used as a long term method to find an article. DOI provides a resolving service, a long term way to find an article, the *same* article, even if it moves or has its metadata changed.

And that costs money, which is why it isn't free. There are other solutions, like Handle.net for example. These systems are incredibly busy. Handle.net serves 70 and 100 million resolution requests per month. The handle.net website has an excellent FAQ about why a persistent way to reference digital content in this manner is a good idea. http://handle.net/faq.html

Regards,
Jason

[Ojser]

The whole premise behind DOI is that you have 'durable' links that do not change.

They are as durable as servers are durable, whatever the identifier method (DOI or nameDateJournal).

They are also not unique (there may be more than one John Smith in the world who has written a paper called 'Widgets', for example).

At the same moment, same journal title too? Very very very unlikely.

Technically, not even traditional URLs can be used as a long term method to find an article. DOI provides a resolving service, a long term way to find an article, the *same* article, even if it moves or has its metadata changed.

Is the durability linked on archive format (DOI or other identifier) or on the server maintenance?
Isn't it on the availability of servers? So, they are long as the servers are maintained.
...

http://handle.net/faq.html

Well, this is already much cheaper than DOI, which is a good alternative.
But, it is still expensive though, especially for small business, and in regard to the real costs.