May 15, 2012 at 12:30 pm
I don't know that the contributors so far have looked at this from a perspective other than as a DBA.
As a DBA I like the idea of being able to spin up a private SQL server in the cloud to perform tests, act as a sandbox or provide data collection for a web service.
As a software developer I can push table creation, indexes, constraints and functions right from Visual Studio without wearing my DBA hat.
As an end user people can use a SQL server with their third party enterprise application to log antivirus, network monitoring, CRM, etc. Many of the apps have their own backup systems and/or the user doesn't care if they lose historical information (i.e.: how many viruses were caught last year).
Microsoft is providing a service that many of us could use including novice SQL admins as well as your typical I.T. person. It's also a nice tool for someone to start with without having to purchase the hardware and the licenses.
May 15, 2012 at 12:54 pm
One of the IMO important advantages of a server in the cloud is that you can move it from one machine to another without anyone even noticing.
May 15, 2012 at 4:34 pm
P Jones (5/15/2012)
It's just the same as the proliferation of Access databases and Excel spreadsheets out there and we have to control those somehow!
While I agree in principle, the proliferation of MS Access has made me a lot of money over the last 15-ish years by people employing me to come in and fix up their "mission critical" Access databases & systems.
May 15, 2012 at 5:26 pm
Steve Jones - SSC Editor (5/15/2012)
How often do we see on this site decisions made by people who work as DBAs that are awful? These people are often trained for this work and still make bad decisions. How can somebody with little or no training come up with something good.
I know a little bit about chemistry. That does not mean that I should go down to the labs and start making something. I would likely blow the building up. I leave chemistry to the trained professionals. Databases and applications should be the same.
I see your point, and lots of time this is true. However people without any formal training, or doing things as a hobby come up with good ideas all the time. Lots of business people do a great job of analysis with data that you or I would never come up with.
We ought to enable some self-service, or some ideas, but limit the impact. The problem isn't someone building an Access app, it's that they build one and it is a good idea, it gets used more and more, without any investment being made to improve it to something that is more scalable or resilient to stress.
EXACTLY! The key to success is to "give them the opportunity to fail". It's a remarkable thing when they don't fail. Let the users innovate! Hell... enable them! They know what they need to have done better than any of us do. Embrace their innovation! Keep in touch with them and help them where you can! When they get something to the point where it becomes mission critical, you'll already have an implementation method in mind because you'll have been part of the whole thing to begin with. Think you don't have the time for such a thing? My friends, it's going to happen with or without you. My recommendation is to get involved so you can help control it because you're not going to be able to stop it. 😉 We applaud innovation within our own ranks. Applaud and embrace the innovation of the users. Be an enabler. 🙂
--Jeff Moden
Change is inevitable... Change for the better is not.
May 15, 2012 at 5:50 pm
I have to lean towards Jeff's view on this one. For example, I've been teaching one of my business users basic querying fundamentals and how to effeciently get data from a massively badly designed vendor system we have.
Enable the end user. Get them thinking from a technical perspective and above all LET THEM GET THEIR HANDS DIRTY! Wha?!
Wait a second, why do we want users mucking about in our systems? Well, short answer, we don't, we want them in THEIR systems mucking about. How many times have you walked into a business meeting, asked for technical clarifications, and had a room full of people look at you like you've just fallen off the boat? Get one of those users who's a 'business type' to understand your dilemma and they'll paraphrase it into the local lingo for you. BAM, off and running!
The second part of the consideration is that they make mistakes, bad design decisions, etc etc etc. Yep, of course they do. They're newbies. Meh, it happens. I've made a pretty good career off that by helping them fix the ones that worked instead of having me constantly on hand for the 20 ideas that didn't. I get paid better by helping them fix the one that makes them money! Systems today can handle the stress of low-impact databases for prototyping/low-end work. Mom and pop up the street who need to keep their inventory and sales totals somewhere safe really don't need to hire me.
I agree that, in general, every Tom Dick and Harry system doesn't need a DBA, or even a Developer. If it's a low impact system, go NUTS with ORM. Have fun with it.
When you reach your limits with that 1 project that's blown out of scope, give me a call... I'm here.
Never stop learning, even if it hurts. Ego bruises are practically mandatory as you learn unless you've never risked enough to make a mistake.
For better assistance in answering your questions[/url] | Forum Netiquette
For index/tuning help, follow these directions.[/url] |Tally Tables[/url]
Twitter: @AnyWayDBA
May 15, 2012 at 6:02 pm
Evil Kraig F (5/15/2012)
When you reach your limits with that 1 project that's blown out of scope, give me a call... I'm here.
+1
May 16, 2012 at 8:57 am
I think a big part of the problem lies in a convergence of false assumptions and the desire to offer a new product. To me, the implied assumption is that self-service is always good and empowering (yada yada yada) and then all critical thought ceases. Then *poof* a new hammer for the masses ,if you will, is put on the market and toes get bruised and broken. If I could assemble an airplane by pressing a few buttons, I would have to ask myself: should I build an airplane? Who will keep it running? Do I want to get in it?
May 16, 2012 at 9:44 am
cdonlan 18448 (5/16/2012)
I think a big part of the problem lies in a convergence of false assumptions and the desire to offer a new product. To me, the implied assumption is that self-service is always good and empowering (yada yada yada) and then all critical thought ceases. Then *poof* a new hammer for the masses ,if you will, is put on the market and toes get bruised and broken. If I could assemble an airplane by pressing a few buttons, I would have to ask myself: should I build an airplane? Who will keep it running? Do I want to get in it?
Very fair questions, but there are people that build airplanes, or ultralights in their garages. Some crash, but the people tackling these ideas sometimes make wonderful products.
It's not an always this way or that way thing. I don't think I implied, or most anyone has implied, that self-service is always good, or it will always turn out good projects or that there are not limits. I think most of the discussion that encourages this has been around give people a chance to experiment and then work with them on the projects that work out well. Shut down the projects that don't, and have some limits that may constrain the growth to some reasonable level.
May 16, 2012 at 9:54 am
Actually I agree with you. My speculation was to how the product even made it to market. Can't you read minds? 🙂 I suppose my own assumption (and fear) is that this new magical self-service database would be used as part of a mission- critical solution. If that's not the case then the risk seems negligible relative to the benefits of 'pushing the boundaries of the technology'. I think somebody posted earlier something about the necessity of education and I think that's a great point as well.
(Sorry Steve, this didn't post where I thought it would. It is a reply to your reply.)
May 16, 2012 at 10:57 am
cdonlan 18448 (5/16/2012)
I suppose my own assumption (and fear) is that this new magical self-service database would be used as part of a mission- critical solution.(Sorry Steve, this didn't post where I thought it would. It is a reply to your reply.)
Not a good assumption, but a genuine fear. Too often items are pushed beyond their limits and become mission critical because no one stops and makes an assessment when the application becomes useful to more than one person
Viewing 10 posts - 16 through 24 (of 24 total)
You must be logged in to reply to this topic. Login to reply