Today's rant is coming between me and a day of skiing so I'll keep it brief. While trying to get to the bottom of why I can't enjoy offline access to Google Apps & other web-based applications with Gears on Snow Leopard I came across a post noting Chrome, Opera to support html5 webdb, FF & IE won’t. This seemed curious as HTML5 is powering on towards last call and there are already multiple implementations of both applications and clients that run them. Here's where we're at:
Mozilla's argument that "we really don't want SQL" is far more troublesome and posts like this follow an increasingly common pattern:
In any case the API [cs]hould be versioned so we can offer alternatives like WebSimpleDB in the future. Right now though the open web is being held back by outdated standards and proprietary offerings controlled by single vendors (e.g. Adobe's AIR and Microsoft's Silverlight) are lining up to fill in the gap. Those suggesting "it's worth stepping back" because "there are other options that should be considered" which "might serve those needs better" would want to take a long, hard look at whether their proposed alternatives are really ready for prime time, or indeed even necessary. To an outsider trying to solve real business problems today a lot of it looks like academic wankery.
Read full history - NoSQL "movement" roadblocks HTML5 WebDB
- Opera: "At opera, we implemented web db [...] it’s likely we will [ship it] as people have built on it"
- Google [Chrome]: "We’ve implemented WebDB … we’re about to ship it"
- Microsoft [IE]: "We don’t think we’ll reasonably be able to ship an interoperable version of WebDB"
- Mozilla [Firefox]: "We’ve talked to a lot of developers, the feedback we got is that we really don’t want SQL [...] I don’t think mozilla plans to ship it."
Mozilla's argument that "we really don't want SQL" is far more troublesome and posts like this follow an increasingly common pattern:
- Someone proposes SQL for something (given we've got 4 decades of experience with it)
- Religious zealots trash talk SQL, offering a dozen or so NoSQL alternatives (all of which are in varying stages of [early] development)
- "My NoSQL db is bigger/better/faster than yours" debate ensues
- Nobody does anything
In any case the API [cs]hould be versioned so we can offer alternatives like WebSimpleDB in the future. Right now though the open web is being held back by outdated standards and proprietary offerings controlled by single vendors (e.g. Adobe's AIR and Microsoft's Silverlight) are lining up to fill in the gap. Those suggesting "it's worth stepping back" because "there are other options that should be considered" which "might serve those needs better" would want to take a long, hard look at whether their proposed alternatives are really ready for prime time, or indeed even necessary. To an outsider trying to solve real business problems today a lot of it looks like academic wankery.
