- From: Gannon Dick <gannon_dick@yahoo.com>
- Date: Wed, 2 May 2012 07:56:25 -0700 (PDT)
- To: "eGov IG \(Public\)" <public-egov-ig@w3.org>
Most definitely so. This is a prequel to "Best Practices for Publishing Linked Data W3C Editor's Draft 29 April 2012"[1]. In the Vocabulary Discovery Checklist, step one is "Specify the domain". This assumes the data.gov.uk model of one "master" domain, and furthermore, that this domain is already set up. The eGov IG (and the hundreds of !UK governments) can't make the "already setup" assumption IMO. The first baby step is to specify the "verticals", to go from the familiar Verticals Org Chart format in a hierarchal form [2] to a linked data-ish representation[3] and back to the Org Chart in table form[4]. The Verticals Org Chart in table form is what it is, but has features that address Governance rather than Government. - You can't put it in "alphabetical order". This reflects the complex nature of domains, but is a bit traumatic for non-techies, and prompts them to wonder "So when did this [SKOS] stop being [S]imple!". Terminology should not confuse newbies, but it does. - There are no domain URI's minted. Names (URN) are retained and the rewrite by regular expression mechanism is the same for URN:LEX[5] as for domain URI's and URL's[6]. - There are no actual linked pages, just references and then only to "official" government sources e.g. the LOC ID servers or existing (Australian) sites. One could add DBpedia, etc.. At this early stage, though, "Domain Security Concerns" are dominated by practitioners of "the old ways". This features xenophobia, with a moderate chance of paranoia. --Gannon [1] https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html [2] http://www.rustprivacy.org/2012/urn-lex/linked.data.is.different.pdf [3]  http://www.rustprivacy.org/2012/urn-lex/linked.data.is.different.xhtml (server links do not work, because the server is not running here) [4]  http://www.rustprivacy.org/2012/urn-lex/linked.data.is.different.html [5] http://tools.ietf.org/html/draft-spinosa-urn-lex-06 [6] http://tools.ietf.org/html/rfc3986#appendix-B ________________________________ From: John Erickson <olyerickson@gmail.com> To: eGov IG (Public) <public-egov-ig@w3.org> Sent: Wednesday, May 2, 2012 7:13 AM Subject: Re: technical web standards for eGov This discussion seems to be less about "technical standards for eGov" and more about "recommended best practices for publishing linked data" --- the latter being within the purview of the W3C Government Linked Data (GLD) Working Group <http://bit.ly/xVncUL> In addition to having recently produced FPWDs on recommended vocabularies, W3C GLD WG has in "Editor's Draft" stage a "Best Practices for Publishing Linked Data" document, intended to aggregate best practices across the government linked data lifecycle, from minting URIs to creating publication policies to persistence and provenance. Whether to adopt specific technologies such as D2R should be considered by government stakeholders in this context... On Wed, May 2, 2012 at 6:20 AM, Paola Di Maio <paola.dimaio@gmail.com> wrote: > Thank you v much G > > sounds great! > >>IThe D2R Server runs in a browser on your local machine, at the "local" domain. > > thats what puzzles me I guess. why locally, and  not virtual, in the cloud, where it should run? > > I presume it is because the data sets needs to be downloaded, normalised and tidied up before being queried to reduce errors? > > I cannot think of any other reason why SW applications would need to run locally. > > P > > All of the files are in the >> right place. Since you can also do an RDF dump, you have everything you >> need to run Callimachus - which is what runs the W3C GLD Community Directory >> http://dir.w3.org/ So, the D2R Server is just another route to Government >> Linked Data. My tool is intended to address the "domain" problem. If the >> domain covers too wide an area you cannot deliver science (Brand knows a lot >> about this) and maintain service levels across the domain. If the sewer >> system in NYC is stressed it makes no difference if the sewer system in Los >> Angeles has extra capacity. Sewer system relationships do not change when >> linked. You can get a visualization of what "stressed" means, but linked >> data does not solve the problems it may uncover. >> >> I'll try and post some tutorials tomorrow. >> >> --Gannon >> >> >> ________________________________ >> From: Paola Di Maio <paola.dimaio@gmail.com> >> To: Brand Niemann <bniemann@cox.net> >> Cc: Gannon Dick <gannon_dick@yahoo.com>; paoladimaio10@googlemail.com; eGov >> IG (Public) <public-egov-ig@w3.org> >> Sent: Tuesday, May 1, 2012 2:25 PM >> >> Subject: Re: technical web standards for eGov >> >> Thank you Gannon and Brand >> >> >> hadnt understood I need to download d2r from github to make sense of >> the ttl file >> >>> I used the D2R Server because 1) It delivers Linked Data in a portable >>> format, and 2) The web server is relatively easy to get working on a local >>> machine. >> >> ok- but can D2R functionalities not be served as a web service? >> (thats what i'd aim for) >> >> Please consider writing some tutorials and how tos >> for people to get on the same page on you with this data, and let me >> know if I can help with foolproofing and promoting any outcomes. >> Or even better, send some diagrams!! >> >> I ll have a go though (feeling inadequate) >> >> Cheers >> >> P >> >> >> >> On Tue, May 1, 2012 at 5:25 PM, Brand Niemann <bniemann@cox.net> wrote: >>> I was a able to do something with Gannon’s data: >>> http://semanticommunity.info/Gannon_Dick >>> >>> >>> >>> From: Gannon Dick [mailto:gannon_dick@yahoo.com] >>> Sent: Tuesday, May 01, 2012 12:02 PM >>> To: paoladimaio10@googlemail.com; eGov IG (Public) >>> Subject: Re: technical web standards for eGov >>> >>> >>> >>> XML (HTML, etc.) is human readable, but linked data is not "human >>> comprehensible". Visualizations provide some ideas, but it's an old story >>> ... when you are looking at a marked-up document you are looking at a >>> picture of text content, not the text itself. DATA.GOV published a Drupal >>> 6 >>> implementation some time ago on github. There is a lot of terminology to >>> cope with there too. >>> >>> >>> >> >> >> >> >> >> You don't need to do any further development of the framework. >>> The D2R Server software is available on github. It is also possible, if >>> you >>> control the data base to extract data directly in a familiar >>> outline/subject >>> heading form. If you don't know what the screen shots mean it is because >>> trying to draw a picture of a hypercube looks strange under the best of >>> circumstances. "What the heck is that ?" is fine as long as you >>> understand >>> that "that" is something desirable to have when you are done. >>> >>> >>> >>> That said, eGov is a "vertical", a Public Utility, of the Top Level >>> Domains. No other formulation makes sense ... the risk is that at the >>> bottom, governance breaks down. At the bottom, linked data and the >>> semantic >>> web break down from the same cause, although it can not be said for the >>> same >>> reason. Commercial domains are free to "fly-over" (exclude) customers >>> from >>> their markets. Governments can not operate this way (excluding people) >>> for >>> very long. >>> >>> >>> >>> I'm sure I need more examples that look like familiar organization tables >>> ... working on it :o) >>> >>> >>> >>> --Gannon >>> >>> >>> >>> ________________________________ >>> >>> From: Paola Di Maio <paola.dimaio@gmail.com> >>> To: eGov IG (Public) <public-egov-ig@w3.org> >>> Sent: Tuesday, May 1, 2012 8:59 AM >>> Subject: technical web standards for eGov >>> >>> >>> Now, having Gannon send a file that the browser cannot open brings up >>> another issue, which may be more general to SW. >>> >>> Should we, as a W3C IG ensure that (at least one version of) data >>> files and docs re e-Gov shared via the web are at a minimum, >>> accessible via web based technologies (I know the browser, is there >>> anything else that is used to navigate the web these days that I may >>> not know of?) and do not require physical data downloads? >>> >>> When people send me dumps expect me to download stuff >>> then they call it SW I cannot honestly believe this is where we are >>> (sorry if I missed some important updates) >>> >>> >>> P >>> >>> >> >> >> > -- John S. Erickson, Ph.D. Director, Web Science Operations Tetherless World Constellation (RPI) <http://tw.rpi.edu> <olyerickson@gmail.com> Twitter & Skype: olyerickson
Received on Wednesday, 2 May 2012 14:57:01 UTC