jump to navigation

Friday Philosophy – Know Your Audience May 7, 2015

Posted by mwidlake in Blogging, Friday Philosophy, Presenting, publications.
Tags: , ,
9 comments

There are some things that are critical for businesses that can be hidden or of little concern to those of us doing a technical job. One of those is knowing who your customers are. It is vital to businesses to know who is buying their products or services. Knowing who is not and never will buy their products is also important (don’t target the uninterested) and knowing and who is not currently buying and might is often sold as the key to ever growing market share and profit. But fundamentally, they need to know who the current customers are, so they can be looked after {I know, some businesses are shocking to current customers, never understood that}.

This should also be a concern to me.

Why? Well, I “sell” something. I don’t charge for it, but I put out my blogs and my tweets and my presentations. I’ve even stepped up to articles. So I am putting a product out there and I want people to use it. Any of us who blog, tweet, facebook or in some way communicate information are fundamentally trying to talk to people. It’s fine to just put stuff out there and see who comes, but if I am doing this in order to reach an audience, well, who is my audience?

I know who my audience is. I’m British. I live in the UK, 75% of my presentations are in the UK, 95% of my work has been in the UK. I drink tea as a hobbie, queue as only the British know how, want my ale at room temperature and I am self-deprecating in my humour. At least, I’d like to think I am, but please forgive me if I fall short of your expectations.

My Audience is UK:

Who comes looking from where

Who comes looking from where

My Audience is American.

Dang!

As you can see from the above, my reasonable assumption was wrong. Those are stats I pulled from my blog about visits by country for a recent period. Most of my audience is in the US. For this particular period the UK is my second highest audience and India is third, but I dug in a little more and at times my Indian audience is higher than my UK audience.

Other countries move up and down but the above graphic is representative – European counties, Canada, South America and Australia all are prominent areas for me, and South Korea – big technology country, South Korea, so I should expect a reasonable showing from there. However, I’ll just let you know that last year (different graph, I hasten to point out) I had only 1 visitor from the Vatican, Vanuatu and Jersey (part of the UK!) each. I’m a bit gutted about Jersey, having worked there once, but the Vatican? Does the Pope need a VLDB?

I have noticed a spike of interest in a given month by a country if I go and present there, but it does not last for long.

What about my Tweet world? The below shows where my followers are from:

Peeps wot Tweets

Peeps wot Tweets

It is nice that this graph emphasises that “others” outside the top 10 are larger source of audience tham any individual country, but it shows a similar pattern to my blog. I’m mostly talking to my American cousins, the home crowd and our friends over in India. I suppose if you think about the number of people working in IT (and, to a lesser extent, just simply living) in countries across the global, the numbers make a lot of sense. If I was doing this analysis on a database of the raw data I’d now be correlating for population size and trying think of a proxy I could use for “IT Aware”.

So now I know who my audience is. Does this mean I should alter the tone of my posts to be more American or International, or is the British flavour of my erudite utterances part of the appeal?

I have noticed one change in my output over that last year or so, as I have become more aware of the geographical spread of my audience. I tend to explain what I recognise as odd phrases (above paragraph allowing) or UK-centric references a little more. And I try to allow for the fact that not everyone visiting my blog speaks English as a first language. But in the end, I have to use the only language I know. However, I don’t think I appreciate well when I am using colloquial phrases or referencing UK-centric culture. I’ll try harder.

One thing I do resist is WordPress trying to auto-correct my spelling to US – despite the fact that the app knows I am in the UK. Maybe I should spend some time trying to see if I can force the use of a UK dictionary on it? I won’t accept corrections to US spelling because, damn it all chaps, English came from this island and I refuse to use a ‘Z’ where it does not belong or drop a ‘u’ where it damned well should be! And pants are underwear, not trousers, you foolish people.

There is another aspect of my blog posts that I find interesting, and it is not about where my audience is – it is about the longevity of posts. Technical posts have a longer shelf life. My top posts are about oddities of the Oracle RDBMS, constantly being found by Google when people are looking at problems. A couple of the highest hitters I put up in 2009 when almost no one came by to look. However, my “Friday Philosophies” hit higher in the popularity stakes when first published but, a month later, no one looks at them anymore. Stuff about user groups and soft skills fall between the two. Some of my early, non technical posts just drifted into the desert with hardly any notice. Sadly, I think a couple of them are the best things I have ever said. Maybe I should republish them?

My First Published Article April 21, 2015

Posted by mwidlake in Presenting, publications, UKOUG.
Tags: , , ,
7 comments

I’ve been blogging now for almost 6 years and presenting at conferences for… 12 years (really? Good grief!). I’ve even written and delivered several courses, ranging from 1 day to 3 days in length. However, up until now I’ve never been what I would term published – ie managed to persuade another organisation or person to publish something I have written.

That changed a few days ago when the latest UKOUG “Oracle Scene” magazine came out, which included the first of a small series of articles I am doing on how the Oracle RDMBS works – the processes and activities that underlie the core RDBMS engine. It’s based on my “how Oracle works in under an hour” presentation where I give the audience an overview of things like the redo mechanism, what a commit *is*, how data is moved into and out of memory and which parts of memory it resides in, how a point-in-time view is maintained… things like that. Many people don’t really know any of this stuff, even skilled and experienced developers and DBAs, as you can get by without knowing it. But understanding the core architecture makes a lot of how oracle works make more sense.

The below is a screen shot of the title and first paragraph, but you can use the link above to see the whole article.

Title and first paragraph of the article

Title and first paragraph of the article

I’m not sure why it has taken me so long to publish something other than via my blog and presentations. I know part of it is the fear of putting something out there that is wrong or misleading. If it is on my blog, heck it’s only a blog and I stick to things I give test cases for or my thoughts and opinions (which are intrinsically open to interpretation). My presentations are certainly put “out there” but again I of course try to ensure what I say I can back up. I think the key thing is that in both cases it is very obvious who you can blame if it turns out I have made a mistake. Me.

But when something is going to be published I feel that (a) it might be taken more seriously so I need to make extra sure it is correct and (b) if I get something wrong or, more concerning, mislead anyone then the people publishing the article could also be put in a poor light. I think that is what has made me wary.

The irony is that the first thing I get published, I know that there are some inaccuracies in there! The article (and also the presentation it is derived from) is an introduction to a lot of technology and I have to simplify things and ignore many exceptions to keep it small and easy to digest. It’s how it works 90% of the time and you need to know that so you will better understand the exceptions and finer detail I don’t have time to tell you about. For the physical presentation I spend a minute at the top of the talk saying I have simplified, occasionally lied, but the overall principles and feel is correct. I had to drop that bit out of the article as, well, it took a lot of words to explain that and the article was long enough already!

Another reason NOT to publish is it takes a lot of time and effort to prepare the material in a way that is polished enough to be printed and I know from friends that the actual financial payback for eg writing a book is very, very, very poor. No one I know makes enough from royalties on technical books to make the effort worth while {though there are other less tangible benefits}. But I have time at present so I can afford to do these things.  If you want to make money out of publishing, write about a load of elves, an often-wimpy trainee wizard or something with sex in. Or all three together.

I did nearly put a technical book together about 10, 12 years ago, called “The Little Book of Very Large Databases” as it was something I knew a lot about but the issues were rarely discussed publicly – most VLDBS were (and are) run by financial organisation or “defence” {why can’t they be honest and refer to themselves as “Killing & Spying”} and they don’t talk. O’Reilly was doing several small, A6 booklet-type-books at the time that it would have suited. I can’t do it now, I know nothing about Cloud and some of the 12C features that would help with VLDBS, so I missed the boat. I regret not giving it a go. However, there is a possibility I might be involved in a book sometime in the future.

I have to thank Brendan Tierney for hassling me into doing this series of articles. I’m not being derogatory when I say he hassled me, he did, but Brendan did so in a very nice way and also gave me the odd toe in the backside when I needed it.

I also have to thank Jonathan Lewis. If this article had been a book he would have got a huge mention for being my technical reviewer. He was good enough to look over the article and let me know a couple of things he felt I had over simplified, some things with the flow and also something I had simply got wrong. You know that bit in books about “thanks to Dave for assisting but all mistakes are mine”. Well, I always thought it was a bit overly… defensive? Well now I don’t.

All mistakes are mine. I want no blame falling on the people who helped me!

I still can't take my Bio too seriously

I still can’t take my Bio too seriously

Oracle documentation on a Kindle January 18, 2012

Posted by mwidlake in publications.
Tags: ,
7 comments

I recently bought myself a Kindle – the keyboard 3G version. Keyboard as I know I will want to add notes to things and the 3G version for no better reason than some vague idea of being able to download things when I am away from my WiFi.

So, how about getting Oracle documentation onto it? You can get the oracle manuals as PDF versions (as opposed to HTML) so I knew it was possible and that others have done so before. A quick web search will show a few people have done this already – one of the best posts is by Robin Moffat.

Anyway, this is my take on it.

1) Don’t download the PDF versions of the manuals and then just copy them onto your kindle. It will work, but is not ideal. PDF files are shown as a full page image in portrait mode and parts are unreadable. Swap to landscape mode and most text becomes legible and you can zoom in. In both modes there is no table of contents and none of the links work between sections. All you can do is step back and forth page by page and skip directly to pages, ie goto page 127. This is not so bad actually as quite often the manual states the page to go to for a particular figure or topic.

2) Do download the MOBI format of the manuals you want, if available. Oracle started producing it’s manuals in Mobi and Epub format last year. I understand that Apple’s .AZW format is based on .MOBI (Mobipocket) format. As such text re-flows to fit the screen of the Kindle. I’ve checked a few of the DBA_type manuals for V10 and V11 and Mobi files seem generally available, but not a couple I checked for 10.1. If there is no Mobi, you can still revert to downloading the PDF version.

3) You cannot download a set of manuals in this format and you won’t see an option to download an actual manual in MOBI format until you go into the HTML version of the document.

I can understand that it would be a task for someone in Oracle to go and create a new downloadable ZIP of all books in these formats or, better still, sets to cover a business function (like all DBA-type books and all developer-type books), but it would be convenient.
Anyhow, go to OTN’s documentation section, pick the version you want and navigate to the online version of the manual.

Here I go to the 11.2 version – note, I’m clicking on the online set of manuals, not the download option.


Select the HTML version of the document you want, in this case I am grabbing a copy of the performance tuning guide. As you can see, this is also where you can choose the PDF version of the manual

Once the first page comes up, you will see the options for PDF, Mobi and Epub versions at the top right of the screen (see below). I wonder how many people have not realised the manuals are now available in new ebook formats, with the option only there once you are in the manual itself?

I’ve already clicked the Mobi link and you can see at the bottom left of the screen-shot, it has already downloaded {I’m using Chrome, BTW}. Over my 4Mb slightly dodgy broadband connection it took a few seconds only.

4) I don’t like the fact that the files are called things like E25789-01.mobi. I rename them as I move them from my download directory to a dedicated directory. You then attach up your kindle to your computer and drag the files over to the kindle’s “documents” folder and, next time you go to the main menu on the kindle, they will appear with the correct title (irrespective of you renaming them or not)

5) If you download the PDFs I would strongly suggest you rename these files before you move them to the kindle as they will come up with that name. I have a booked called e26088 on my kindle now – which manual is that? {don’t tell me, I know}. I have not tried renaming the file on the kindle itself yet.

6) You don’t have to use a PC as an intermediate staging area, you can directly download the manuals to your kindle, if you have a WiFi connection. Go check out chapter 6 of the kindle user guide 4th edition for details, but you can surf the web on your kindle. Press HOME, then MENU and go down to EXPERIMENTAL. click on “Launch Browser” (if you don’t have wireless turned on, you should get prompted). I’d recommend you flick the kindle into landscape mode for this next bit and don’t expect lightning fast response. If it does not take you to the BOOKMARKS page, use the menu button to get there and I’d suggest you do a google search for OTN to get to the site. Once there navigate as described before. When you click on the .Mobi file it should be downloaded to your kindle in a few seconds. Don’t leave the page until it has downloaded as otherwise the download will fail.

There you go, you can build up whatever set of oracle manuals you like on your ebook or kindle and never be parted from them. Even on holiday…

I’ve obviously only just got going with my Kindle. I have to say, reading manuals on it is not my ideal way of reading such material. {story books I am fine with}. I find panning around tables and diagrams is a bit clunky and the Kindle is not recognising the existence of chapters in the Oracle Mobi manuals, or pages for that matter. However, the table of contents works, as do links, so it is reasonably easy to move around the manual. Up until now I’ve carried around a set of Oracle manuals as an unzipped copy of the html download save to a micro-USB stick but some sites do not allow foreign USB drives to be used. I think I prefer reading manuals on my netbook to the kindle, but the kindle is very light and convenient. If I ever get one of those modern smart-phone doo-dahs, I can see me dropping the netbook in favour of the smartphone and this kindle.

Of course, nothing beats a big desk and a load of manuals and reference books scattered across it, open at relevant places, plus maybe some more stuff on an LCD screen.

Want to Know More about Oracle’s Core? October 19, 2011

Posted by mwidlake in performance, Private Life, publications.
Tags: , , ,
14 comments

I had a real treat this summer during my “time off” in that I got to review Jonathan Lewis’s up-coming new book. I think it’s going to be a great book. If you want to know how Oracle actually holds it’s data in memory, how it finds records already in the cache and how it manages to control everything so that all that committing and read consistency really works, it will be the book for you.

{Update, Jonathan has confirmed that, unexpected hiccups aside, Oracle Core: Essential Internals for DBAs and Developers should be available from October 24, 2011}

{Thanks to Mike Cox, who let me know it is already available to be reserved at Amazon}

Jonathan got in touch with me around mid-May to say he was working on the draft of his new book, one that would cover “how does Oracle work”, the core mechanics. Would I be willing to be one of his reviewers? Before anyone comments that there is not likely to be much about core Oracle that I know and Jonathan does not, he did point out that he had already lined up someone to be his technical reviewer, ie someone he expected to know as much as he and help spot actual errors. The technical reviewer is the most excellent Tanel Poder, who posted a little mention of it a couple of months back.

I was to act more like a typical reader – someone who knew the basics and wanted to learn more. I would be more likely to spot things he had assumed we all know but don’t, or bits that did not clearly explain the point if you did not already know the answer. ie an incomplete geek. I figured I could manage that :-).

It was a lot harder work than I expected and I have to confess I struggled to supply back feedback as quickly as Jonathan wanted it – I was not working but I was very busy {and he maybe did not poke me with a sharp stick for feedback soon enough}. As anybody who has had to review code specifications or design documents will probably appreciate, you don’t just read stuff when you review it, you try and consider if all the information is there, can it be misunderstood and, if you find that you don’t understand a section, you need to work out if the fault is with you, with the way it is written or with what is written. When I read a technical {or scientific} document and I do not fully understand it, I usually leave it a day, re-read it and if it still seems opaque, I just move on. In this case I could not do that, I had to ensure I understood it or else tell Jonathan why I thought I did not understand it. If there are sections in the end book that people find confusing, I’ll feel I let Jonathan down.

Just as tricky, on the one hand, as I’ve been using Oracle for so long and I do know quite a lot about Oracle {although clearly not enough in the eyes of the author :-) } I had to try and “not know” stuff to be able to decide if something was missing. On the other, when I wanted to know more about something was I just being a bit too nerdy? I swung more towards the opinion that if I wanted to know more, others would too.

I have to say that I really enjoyed the experience and I learnt a lot. I think it might change how I read technical books a little. I would run through each chapter once to get the feel of it all and then re-read it properly, constantly checking things in both version 11 and 10 of Oracle as I read the drafts and would not let myself skip over anything until I felt I really understood it. As an example, I’ve never dug into internal locks, latches and mutexes much before and now that I’ve had to learn more to review the book, I have a much better appreciation of some issues I’ve seen in the wild.

Keep an eye out for the book, it should be available by the end of this year and be called something like “Oracle Core” {I’ll check with Jonathan and update this}. I won’t say it will be an easy read – though hopefully a little easier as a result of my input – as understanding things always takes some skull work. But it will certainly be a rewarding read and packed full of information and knowledge.

Follow

Get every new post delivered to your Inbox.

Join 199 other followers