I had a brief conversation on Twitter about the Microsoft Connect bug reporting system, and received the tweet pictured above. It notes: “I like how you used ‘Connect works’ in that sentence. Great work of fiction.” That made me laugh because it does sometimes seem that Connect doesn’t work all that well for those of us that use it.
Connect was supposed to be a place where users could report bugs and make suggestions. Some people file a lot of bugs and suggestions, though more than a few seem slightly silly. Others file what appear to be support requests, and I’ve had a number of documentation notes. There are even Connect items filed on Connect itself, though I can’t understand why this particular one is “postponed”. There are some items that have been open for years, even with hundreds of votes.
The Connect system is supposed to feed directly to product groups and there are more than a few times I’ve received feedback through the system from developers I know personally that provide comments. I know developers are seeing the items, but there appear to be a few times of year when the SQL Server group mass closes lots of Connect items, often without any feedback. The twitter conversation I had above included one of these items.
I remember being excited when Connect was introduced, thinking this would be a good way to get feedback to the product groups, and hopefully influence the managers to prioritize some of the common requests. However that excitement was tempered with the idea that so many items would be submitted, and it would be hard to triage and rank them. Indeed, I rarely find something in the searches I make before submitting, even I’ll have submissions closed as duplicate. The number of items submitted is so high, it’s hard to even comprehend how any group works with the submissions.
With the state of Connect, the lack of feedback, and the mass closures, I’m not sure how well Connect is working these days. However it’s not all bad; I did submit one item that received enough votes to actually make a change in MS policy. Outside of that, I think @SirSQL’s statement might be closer to the truth than I would hope.
The Voice of the DBA Podcasts
We publish three versions of the podcast each day for you to enjoy.