> When it comes to SEO I think you'll find this link interesting: > http://code.google.com/intl/de-DE/web/ajaxcrawling/docs/getting-started.html
I have come across this page before and it looks like a whole lot of rules and processes that need to be in place in order to support ajax crawlable sites. Does struts2 "just work" for ajax crawlability or does the team need to follow strict guidelines and practices? Tapestry simply works out of the box. Since the ajax fragments are known on the server, it can decide whether to use partial page (XHR + javascript) or full page update as part of the framework. On Thursday, 26 January 2012, Christian Grobmeier <grobme...@gmail.com> wrote: > Please don't get me wrong, there are many things in which Tapestry > beats other frameworks. Not only Struts. Look at the hyped Wicket > framework. I have used it and can say I really dislike it. It is > utterly complex and the docs are the worst thing I have ever seen. > With Tapestry I could create a simple CRUD in 2 hours, without any > knowledge. > > When it comes to SEO I think you'll find this link interesting: > http://code.google.com/intl/de-DE/web/ajaxcrawling/docs/getting-started.html > > I expect my users to have JS on these days. And if I need JS I do a > webapplication, not just a website. Websites can be easily optimized > with Struts2. Webapplications - well do I need or even want to crawl > calculations or some other stuff? In my case not. > > Anyway I fully agree that the frontend technology used by Tapestry > (components, yeah) is ways better then what is currently offered by > Struts. > > But one should not say easily framework x is dead, just because > something is here which is better. > > If it comes to ammo: > - T5 has excellent docs (compared to Wicket) > - T5 is so easy to learn (compared to Wicket) as Struts 2 > - T5 supports view components (S2 does not, or not out of the box) > > My personal killer feature: class reloading. Can only be done with > JRebel in S2. It saves much time, and time is money. Explain your > managers every deployment costs 4 minutes, you need to do 15 a day. > > Cheers > > > On Thu, Jan 26, 2012 at 11:44 AM, Lance Java <lance.j...@googlemail.com> wrote: >> One killer argument is SEO (search engine optimisation). Is your webapp >> publicly available and do you want to appear high in google results? >> >> Whilst google's algorithms are hidden, it is known that if searching for a >> ferarri, a site like >> >> www.mysite.com/cars/ferarri >> >> will appear higher in the search results for a site like >> >> wwww.mysite.com/cars?make=ferarri >> >> I'm not familiar with struts2 but if you are returning json, chances are >> that a lot of the site is invisible to the google crawler which does not >> execute javascript. >> >> Tapestry first draws pages with HTML (all links are simple a href=). It >> then decorates the page with ajax if javascript is available. This has the >> benefit that it gracefully degrades for clients with javascript disabled >> (think screen readers and the google crawler). If you have designed your >> tapestry pages correctly to return null for ajax actions where isXhr() is >> false (see taha's @XHR annotation) your pages should gracefully degrade. >> >> I'm not sure about struts2 but I know many sites will simply stop working >> and may simply display a blank page if javascript is disabled (I think this >> might be the case with GWT). Perhaps you could try turning off javascript >> in the browser then viewing each of your POC sites and get a feel for what >> a crawler will see. I know there are tools to help with SEO out there. >> >> Lance >> >> On Thursday, 26 January 2012, Christian Grobmeier <grobme...@gmail.com> >> wrote: >>> Hey guys, >>> >>> while I agree that T5 is amazing I must say Struts2 is not so bad as it is >>> discussed here. There are far more bad frameworks out there. At least S2 >> is >>> not dead, it is actively maintained and developed. Anyway - T5 does have >>> some pretty cool features (class reloading, it is amazing, or components) >>> which I really love. On Struts side I enjoy the easiness: make up an >> action >>> and return json. This way you have more a service layer with Struts and >> can >>> do whatever you want with JavaScript on the frontend. >>> >>> Again, I agree T5 is a great framework. >>> >>> Cheers >>> Christian >>> >>> On Thu, Jan 26, 2012 at 10:58 AM, Dmitriy Vsekhvalnov < >>> dvsekhval...@gmail.com> wrote: >>> >>>> Seriously? Struts vs. T5? >>>> >>>> Go look to market which people you can find more easily & cheaper. At the >>>> very end someone have to support the app. >>>> >>>> Personally i will never go to pick-up struts job because it is dead. >>>> Unless you pay me significant more for my wasted time :) >>>> >>>> T5 simply the best web-framework in java world. So if you tied to java - >>>> go with it. If not.. you know :) >>>> >>>> >>>> >>>> >>>> On Thu, Jan 26, 2012 at 12:50 PM, Thim Anneesens < t.anneess...@ictjob.be >>>wrote: >>>> >>>>> Thanks for the ammo guys ;). >>>>> >>>>> Thim. >>>>> >>>>> >>>>> On 01/26/2012 10:20 AM, Kalle Korhonen wrote: >>>>> >>>>> (Thim, you don't know what poor English is...) It's always difficult >>>>> to win these arguments on technical merits alone, especially because >>>>> they are often looked at one-by-one instead of as a whole. If at all >>>>> possible, try to find an angle that your organization or your manager >>>>> deeply cares about. For example, if you had more people with Tapestry >>>>> experience than Struts people, that'd be a winning argument for me if >>>>> I was a manager. If you can't find anything else, try this: Struts is >>>>> a dying architecture, as proven by these graphs: >>
