'Just Google It'

  • Comments posted to this topic are about the item 'Just Google It'

  • 99% of the internet is crap, but I cant live without the internet.

    We just have to learn how to find out what is crap and what is not.

  • Well this is not so much a rant against search engines, or even programming forums, as it is about the number of people offering bad advise in those forums. I come across it all the time and especially when it comes to database issues.

    As to googling for it, well it has always been the case that you can not know everything about any technology or platform - and if you try then what you know is out of date before you finish. So the real skill has been to be able to find out what you need to know when you need to know it. Of course, being able to understand and use that knowledge comes next. So it does not matter whether it is Google results or an IBM manual, you have to be able to recognise the real meat when you see it.

  • I find Google is great for solving very narrow and well understood problems, or finding direction for broad less understood problems.

    However it's not great for solving broad less understood problems.

  • Isn't it usually unwise to apply a "solution" without first understanding why it might be a solution? That's no different between information found on the web and that found in a book - except that in a book the author probably had time to spell out some of the whys and why nots, making it easier to see whether the proposed solution is likely to be the right one for the problem in front of us.

    If that's right then is the best way to use the Google results to make them the starting point for research, rather than the end of it? And if we're trying to help someone find answers should we also try to find time to suggest why our proposal may help, or where to look for understanding?



  • I agree with David about understanding WHY something may solve a given problem, though of course there are often times when dealing with something new that it's easier said than done.

    Personally I think if you can't understand it fully then the critical thing first of all is to understand the impact of the change. If I don't understand why it'll work, but it's something I can easily change back afterwards or which will have no negative effect on anything then I may just give it a punt and see what happens. If I find several possibly solutions, I'll work out which of them have the least impact (eg requiring a reboot, effecting other software, taking any services currently online offline etc), and then work through them from the least impacting onwards.

    Finally, if I'm really hesitant about a suggested solution (I personally don't like making ANY changes I can't roll back if I can help it), then look into the history and reputation of the person offering the advice, do you get the impression that the person is very knowledgeable in the subject, or just a serial poster. Also, can you find any other people suggesting the same solution (while not being a copy and paste job) on other sites, with respondents confirming the fix worked for them?

    At the end of the day it's really just another area of risk management like we deal with on a regular basis, especially those of us from a Sysadmin background. You can't always be 100% risk free, but the true test is to mitigate as many of those risks as you can before jumping onto the first solution you find.

    Keith Langmead

  • I always try to find at least 3 people recommending a particular solution before I consider it a possibble fix. And then I check with the experts I've found in the forums. Too many people seem to throw out answers trying to get points for giving the right answer without understanding the problem. Also always test any possible solution before applying to production server, every solution is not perfect for every environment, there are just too many variables. If it's a SQL issue I always look at these forums before Google, better chance of getting a good solution.

  • Something that (used to be) a requirement for journalists also applies here: confirm the answer from multiple sources before proceeding!

  • I have to agree with what others have said:

    1) Understand the solution and why it works, if you can (google more if you can't!)

    2) Try to find multiple solutions or confirmations for one of the solutions, especially if you can't fully grasp why the solution should work.

    I should also say that as I've gotten better at SQL (or any programming system really) I find that the move from novice to advanced is not just about learning the solutions to common problems, but about finding tools that help trace new problems faster.

  • While I appreciate the point of view of the article, I think it misses two points:

    First, there's a great ecosystem of what I call "cargo cult" developers and adminstrators which I will sum up as "I can haz teh codez plz". If you're one of them, best wishes as you work down the list of google hits for your query. If you're not one of them, surely you're aware they exist, and you will apply caveat emptor (and of course, search results are free and you often get what you pay for) to your top search result returns.

    Secondly, a tangent to "you get what you pay for". What is the alternative to free search? Probably finding a consultant out there who understands the specific issue, arranging a maintenance contract or adhoc service call, or that you spend time and money over years getting them to understand your domain in detail. Or, developing the skills in question yourself or in your organization. Getting to this endpoint is expensive in both time and money. I assert that it can be cost effective (in both time and money) to attempt to synthesize the results of internet search on related topics, and go down a few rabbit-trails and dead ends in pursuit of the true solution.

    Finally, I'll mention that the only cookbook-solutions from the internet I forward to our networking team is MS KB articles, which usually are pertinent to the issue if they return from a search. I must say I find it highly ironic that putting an obscure SQL Server error message verbatim in Google gets pertinent MS KB articles with much greater accuracy than Bing.

  • You have to always keep your 'head' and whenever possible verify all aspects and compare them to your situation. More often than not things that might seem relevant are completely irrelevant to your situation.

    So true. My situation sometimes is that I'll get a SQL or Windows error that I can't solve on my own. I'll plug the error text or # into a search engine and get a bunch of hits. The advice given in these results can lead you down 3 or 4 different tangents leaving more confused than than when you started :-D. It's worth a little investigation just to see which avenue(s) might be relevant to your case.

    And I agree with Keith that Google yields better results for these types of queries than Bing.


  • Good article. Kinda reminds me of people who blindly follow their GPS -- the device says "turn left here," even though it'll take them right into a lake.

    Like everything else, Google is a tool. People need to know that a tool helps them do work -- it doesn't do the work for them!

    Check out my blog at https://pianorayk.wordpress.com/

  • When reading information results from ANY search engine online just remember to turn your "common sense filter" on. and you will be ok. Answers are not always accurate or complete for that matter on these search engines. This is kind of a no-brainer for most of us. 😀

    "Technology is a weird thing. It brings you great gifts with one hand, and it stabs you in the back with the other. ...:-D"

  • I'm a moderator on a developer forum.

    It's scary how many people ask how to install a shin brace for their gun so they can hit their foot with 100% accuracy.

    I've lost count of the number of times I've refused to answer a question until the poster proved they weren't asking how to do something stupid and/or harmful.

    I'm more tactful than that, but that's the gist of the initial response. 😉

    Every once in awhile the poster has a legitimate reason to do what appears to be a darn fool thing to be doing. I've found that the ones who do don't mind being questioned first.

  • I almost couldn't do my job without Google. Every week I am finding some problem that I can solve in a matter of minutes/maybe an hour or less simply by entering the error text detail into Google and doing a search.

    However, this assumes that 1) you know the basics of your application (in our case SQL Server) and you only google the weird errors and 2) you know how to filter out junk. Cardinal rule is to search for responses that have *.microsoft.com as the web address, as those most often will be best for the actual answer. But yeah, though for novice DBAs it may be risky, I almost don't think advanced DBAs could do without it at this point, there is so much out there that we don't know and is not in BOL.

Viewing 15 posts - 1 through 14 (of 14 total)

You must be logged in to reply to this topic. Login to reply