Andy Seaborne wrote: > On 09/10/11 21:14, Ian Dickinson wrote: >> On 08/10/11 14:44, Dave wrote: >>> On Fri, 2011-10-07 at 22:22 +0100, Paolo wrote: >>> >>>> Wouldn't be easier to provide a JenaHelloWorld example for people to >>>> checkout/download. >>>> >>>> Then the tutorial becomes: >>>> >>>> svn co >>>> http://svn.apache.org/repos/asf/incubator/jena/Jena2/jena-example/trunk >>>> jena-example >>>> or download jena-example.zip from somewhere >>>> >>>> Eclipse> Import> Existing Project into Worskspace >>> >>> No it doesn't. >>> >>> The value of that tutorial is that it helps set up Jena for use in your >>> Eclipse. So setting up Jena as a User Library, understanding how to >>> create a project etc are important. Just downloading a prebuilt project >>> wouldn't cut it. >>> >>>> I find tutorials with a lot of screen shots annoying if they have too >>>> many screen shots. (And they are expensive to maintain for the editor) >>>> If there is a short-cut, I prefer that than a long sequence of press >>>> button here and there. >>> >>> With respect, you are not the target audience! >> +1 >> >> Paolo - I wrote that article in response to a series of requests >> privately and on jena-dev for some "how on earth do I get started?" >> requests from very inexperienced users. It is, by at least an order of >> magnitude, the most popular item on my blob & website combined. >> >> I actually think we need a lot more hand-holding getting started >> material. I'm actually quite keen to write some, just lack time atm! > > +1 > > As a general point, semweb is currently raw technology. Making that > technology easier to use is of more benefit to "making it happen" than > more complexity.
Yes, easier to use is great. I am a advocate of ease to use software, I think short "getting started guides" are great to start with and I appreciate the recent efforts in providing users with a better experience in getting started with Jena. It's great when you can download, unzip, load your data and run! Or, checkout, package, load your data and run. However, I would argue scalability is also very important. When you put something easy to use in the hands of people, they want to use it for real and for their data. Sometimes their data is quite a lot (and dataset sizes are growing all the time). Monday morning, how do I load 1 billion triples|quads into TDB and serve them with Fuseki? How much RAM do I need? Ease of use here is necessary, but not sufficient. :-) Paolo > > Andy > >> >> Ian >> >> >
