We’re conscious that a number of suggestions have been made about how to develop the front end site and we have been working on some of the smaller items, and are now considering the next phase of wider developments.
The following functionality has been rolled out to the site and we hope will improve your user experience:
- Pagination - when a search is completed the number of returns is now displayed above the results along with options to navigate between the results pages (next, previous, first, last) at the returns footer e.g. http://data.hmg.gov.uk/data/list?keyword=education. We’re conscious that with over 1,100 data sets available navigation between them is critical.
- It’s also worth highlighting here that the entire list of available data can be viewed by undertaking a ‘blank’ search (we’ll be working on ways of categorising the data more efficiently as well – but more of that later on)
- Application lists - All application, idea and dataset url’s are now hyperlinked when either viewing individual items or complete lists.
- Confirmation screen – whenever you contribute to the site through commenting, adding an idea or application you’ll now see a confirmation screen so you know you comments have been successfully received. All submissions to the site are pre-moderated by one of the project team so do bear with us.
The features that we are already considering for the coming releases include:
- Search – extending the searched fields to include the full description, author / submitter, date listed and data file format.
- Data browsing – ability to navigate through the data sets by topic area, date listed, a-z listing, etc
- Meta-data display – We are looking at what appears on each data page and how we can make it easier to navigate.
- Tag Cloud – ability to navigate visually by the tags against each package
- Bringing out our linked data - Integrating SPARQL endpoints within the main site around themes such as education, crime, agriculture, environment, finance, transport. As well as including a list of stores an overview of what is available within each and some SPARQL tutorials in our wiki.
- Wiki – Putting in place a wiki to allow discussion and collaboration across the community, including a page for each dataset.
- Community Membership – possibly through Open ID access or otherwise
Please let us know what you think - including anything that you think that we have missed out - and use the commenting functionality below or email list to give us your thoughts and feedback.
38 comments
Comment by Anonymous posted on
I've just had an account made so apologies if these suggestions have been brought up already.
The first thing I noticed looking at some packages is that the URLs have GET parameters. For example,
http://data.hmg.gov.uk/data/package/?package=uk-tide-times-heights
Should probably just be
http://data.hmg.gov.uk/data/package/uk-tide-times-heights/
This is curious since CKAN itself has clean URLs. This is important because the URLs of the packages would be expected to be long-term stable, and the presence of the question mark suggests (perhaps wrongly, but still) volatility.
This brings me to my next suggestion which pertains to annotations. I would like to be able to make annotations about packages. Perhaps to keep notes as I research, perhaps to augment the available information in a publicly available way. There are many reasons (cf. "Semantic Web" link on the right hand side of the page).
It is minimally enough that the URLs are stable, but even better would be a link in the page that would point to an RDF representation of the information. This would be a nice feature to contribute back to CKAN as well.
This last would go a long way towards making the information on data.gov.uk machine readable and reuseable. There's a video in the FAQ of Sir Tim explaining how this kind of practice is a Good Thing (tm).
Cheers,
-w
Comment by Nathan Surendran posted on
Can I just make a plea for us to use the MindTouch Core open source wiki platform rather than MediaWiki. It's much more in line with what we're trying to achieve here, as it is a platform built for data mashups and collaboration that runs many sites such as the Mozilla developer centre https://developer.mozilla.org.
The underlying platform is called Dream, and is a Web-Oriented Architecture Framework (WOAF) for developing lightweight, highly decoupled web-services. The framework includes of a REST microserver and classes for web communication, XML processing, and writing highly asynchronous code. It is built to make it easier to created compelling new Web 2.0 services and APIs. Many of common cases are taken care of by the runtime implicitly, processing is done concurrently, and interactions with XML and other services is simplified by the Dream library.
MindTouch Dream runs on Microsoft .NET 2.0 and Novell Mono 1.9.1. Dream is the foundation of MindTouch Core. Learn more about MindTouch and Dream:
http://www.mindtouch.com/resources/technology
http://developer.mindtouch.com/Dream
http://developer.mindtouch.com/Dream/About
I'd be very happy to assist in helping with any questions you might have regarding the platform.
Nathan Surendran
http://www.schemaconsulting.co.uk
Comment by Anonymous posted on
That sounds like a terrible idea.
Swapping a popular, well supported, cross platform wiki for one that offers no discernible benefits other than a few buzzwords and a REST server (as if REST requires a whole server stack) sounds like the wrong decision for a government department to be making.
Comment by Anonymous posted on
Hi,
Trying out the SPARQL query function: it seems reluctant to accept a revised query once one has been submitted.
Tried to log in as a new user: it claimed that "richard@light.demon.co.uk is not a valid e-mail address". Is this because the validation is faulty, or because this facility is not available to everyone?
Richard Light
Comment by Anonymous posted on
Specifically, I would suggest
1. A place where people can discuss how to use the data to answer useful questions. e.g. What is the relationship between class size and educational outcomes.
2. Allow anyone (wiki style) to contribute their own data.
e.g. a school might introduce healthy meal options in the canteen. It would be great if they could contribute data on how this might have affected their performance.
Comment by Anonymous posted on
The the red/black colours on the top menu are really harsh on the eyes, Red isn't good in HCI - especially not where it is the main focus of the screen.
The spell checker for reply form is US English! - I certainly don't plan to spell colour without a 'U' 🙂
Bridget, UK
Comment by Anonymous posted on
Search facilities need to be much improved before this site becomes useful - simple key word search is totally inadequate. Any serious practitioner would use an advanced search tool (such as Google Advanced Search) and use a domain filter. Additionally, when adequate search facilities are provided, options to list just titles and change the number of results displayed per page should be added.
Comment by Anonymous posted on
Could you possibly put datasets under sub-headings rather than just 'List All Datasets' please? It makes it impossible to browse usefully. Also for non-technocrats the layout is confusing and much of the content and headings (SPARQL, Apps, Wiki, tag clouds, Digital Engagement Twitter Stream etc.) is meaningless and therefore makes the site difficult and slow to use.
I love the idea, but at the moment it's a bit of a nightmare for the casual user.
Thanks!
Comment by Anonymous posted on
It's very easy to accidentally click on the rating stars. Must make their validity rather suspect.
Comment by Anonymous posted on
Given that this site is promoted by Tim Berners-Lee how about setting a good example and making it pass W3C validation...
http://validator.w3.org/
Comment by Anonymous posted on
Shadbolt and others, government are opening up data for reuse.
Right at the start, your grammar is awful. (Example above).
Try harder please.
Bob Milton
Comment by Anonymous posted on
Tried looking at some basic waste management data (Recyling rates, waste-per-head) to get a feel of how the system works. Recycling link points to a DEFRA page that no longer exists, second example doesn't have a link page at all.
Did nobody do such basic URL checking before launch ?
Comment by Anonymous posted on
On the Recent Apps you have 'UK House Prices'. As the App only covers England it cannot really be called 'UK'.
Tried reporting this via Contact Us but CATPCHA fails every time.
Comment by Anonymous posted on
Searched for tenant satisfaction and got caught in a loop - click on item for NI 160 local authority tenant satisfaction and click on download and you are taken back to the search page. Captcha did not work so could not submit a comment.
Not a good start
Comment by Anonymous posted on
I need to find out the number of homes in Wandsworth that have two or more bedrooms. I've performed key word searches on 'two bedrooms wandsworth' = no return. Then I tried 'population profile london' to widen the net and see if I could drill down from there, and got a return of binge drinking statistics! I'm not expecting perfection but I'll probably can this site now, never to return. Google will do a better job!
Comment by Anonymous posted on
When you click on an external link, it would be good if it opened a new tab, so you can carry on browsing this site.
Comment by Anonymous posted on
Could we please have an improved browse functionality, beyond a simple list? Some sort of hierarchical structure by general subject area, or a graphical representation, would be good. This would enable a better overview to be gained of the many valuable resources that are now available, rather than hit-and-miss searching, or scrolling through one long list.
That said, an excellent initiative; well done; bring on the OS mapping data!
Comment by Anonymous posted on
Would like to be able to search crime statistics by geographical area / address.
Comment by Anonymous posted on
I haven't looked around the site much but the one download I saw was in XLS format. Presumably Microsoft Excel? It would be better to use open formats rather than closed proprietary ones (especially M$).
Comment by Anonymous posted on
Hi, great idea for site and looks good - however, I found it difficult to organise my search results as there are not yet any ways to do that - I would suggest introducing columns to enable ordering search results by number of views / downloads, date added, data year(s), department / owner etc.
Since searching for data sets is the core function of the site, both user friendliness and usefulness would be much improved by this simple enhancement
Comment by Anonymous posted on
Download URLs should be direct, not to some other website - save that for the more information URLs.
Data should be in open formats - not Microsoft proprietory formats.
Comment by Anonymous posted on
A good starting point but needs to go much further.
Searched a number of different datasets and all that I tried had data only down to local authority district level (or equivalent size). If you're wanting the data to be useful to individuals/ communities to make decisions/ target priorities, data needs to be made available to smaller geographical scales (i.e. lower super output areas as a minimum).
Probably early teething troubles, but I also found the system pretty slow in responding.
Comment by Anonymous posted on
It would be useful to have a 'more information' type link prominently displayed on the top level SPARQL page to the less prominent but very helpful SPARQL page in the wiki.
Comment by Anonymous posted on
For your server/hosting team...
Use cache-control headers for static site content such as your images, style sheets etc.
Perhaps review the Varnish reverse proxy & Apache HTTPD configs to ensure you don't leak too much data about your infrastructure.. I've seen at least two different server headers.
http://projects.webappsec.org/Information-Leakage
Server: Apache/2.2.11 (Ubuntu) mod_ssl/2.2.11 OpenSSL/0.9.8g
Apache/2.2.11 (Ubuntu) PHP/5.2.6-3ubuntu4.5 with Suhosin-Patch mod_ssl/2.2.11 OpenSSL/0.9.8g
X-Varnish: 481636068 481483054
Via: 1.1 varnish
Regards
A.
Comment by Anonymous posted on
How disappointing. Only for the computer super-literate, not for ordinary people. What does "Integrating SPARQL endpoints..." mean, for crying out loud.
Comment by Anonymous posted on
Titles don't always refer to country specific data - if it is. Could this be rectified and add countries to the tags.
Comment by Anonymous posted on
What does 'SPARQL' mean? I don't think it is a very helpful menu heading... would be better to have something that people understand
Comment by Anonymous posted on
It's a query language for RDF. I agree it's not helpful to the average punter but then it would make perfect sense to the people who are capable of doing anything with the data - the progammer / webmaster types. Others would simply ignore it.
I can't think of what you could replace SPARQL with What would you say in its place?
Steve
Comment by Anonymous posted on
Actually, having eventually found the "add new comment" link, I must say that it's quite well engineered, with an easy editing and preview facility. Its main drawbacks are the non-intuitive facilities offered by raw HTML, and (esp for a UK Govt initiative) that it only accepts American spelling.
Comment by Anonymous posted on
The applications page could do with some additional navigational features, particularly categorisation of apps and perhaps browsing by rating, popularity and date of submission/last update. Apps are also currently listed over several pages. Showing more apps on one page would improve visibility for those that aren't in the first couple of pages.
Comment by Anonymous posted on
Ever curious about how much other people earn I looked at The Dental Earnings and Expenses report.
I realised that the date of release was 2009-08-04.
What does this mean? 8th April or 4th August? I think we should be told,
AMCT
Comment by Anonymous posted on
Why is there a need to log in, at all? If that log-in is, say, only for system administrators, it should say so. If it's for Joe Public, why? And how can one log in, if there's no facility for registration??
Even this "add a new comment" facility took some searching - eventually I found a link, in tiny font, in an odd location.
It would also be helpful in deciding whether to download an item, if its approx size were indicated.
Comment by Anonymous posted on
I couldn't find anywhere to make the following general comment, so that that absence of a facility is now the subject of a complaint about the front end.
What i wanted to say, if I'd been able, was that some data collected at public expense is currently excluded from public use, paradoxically because it has value (is only useless data to be made available?). For example, all the detail in layers of OS mapping data is currently a money-spinner for the OS (whose management is doubtless licking its lips at the prospect of privatising our asset), but at a much greater cost of lost opportunity for the wider community. I was going to suggest a half-way house between that closed box and the open free-for-all which might benefit US or Chinese GIS more than its UK owners. The half-way house being free data for UK citizens and UK-owned institutions (or EU citizens and institutions) but market-valued cost to the rest of the world.
Comment by Anonymous posted on
Is there any resource on the site for, people such as myself, who would be interested in developing apps. There is seemingly no guidance on connecting to/consuming any of these data services. Or have I missed something obvious?
Comment by Anonymous posted on
How do I get a Username and Password - I cannot find any clues on the pages.
Is it restricted access?
Comment by Anonymous posted on
Hi,
I have just briefly looked over your website, I think the idea of a tag cloud will go down well. Aswell as a calender feature so the users can see what days there have been posts. It just makes it a little bit easier for us all.
In terms of the colours, I have you just use green more throughout the website, maybe replace your current orange them with the green, it represents a fresh look as well as being environmentally friendly.
Best Regards,
Paul Allen
Comment by Anonymous posted on
How about doing some interviews with politicians? You could simply create an youtube channel and post these interviews there. This would have the benefit that I could easily convert them to mp3 (e.g. with youtube mp3) and listen to them when I am on my train to work. It would probably have made the last election easier for me.
Comment by Anonymous posted on
There seems to be no obvious place in which to make this comment which concerns accuracy in the use of our language. Could close attention be paid to this so that an official government site does not set a poor example? For example, there is used the phrase 'outside of government'. This is a tautology. The 'of' should not be there. 'outside government' is all that is needed. There are many similar linguistic errors.