DevCentral

When SOAP has failed.

Don MacVittie — Thu, 17 Jul 2008 19:27:22 GMT

Interesting read over at The Register that posits the question "will Flex SOAP 1.1 work with .NET betas?"

The steps he went through and how he got things going was mildly interesting, more interesting to me was that he even had to ask.

I cannot say how happy I am that the answer was a resounding "Yes, they work together", because if we ever see a "no, these two mainstream Web Services products don't inter-operate" (ignoring those first few painful versions of yesteryear when people were figuring it out), then SOAP has well and truly failed.

The entire purpose of SOAP is interoperability, we shouldn't need to ask these questions. Indeed, they shouldn't even occur to us and the environment should be such that if we try and they don't work together, we assume that we did something wrong.

But some vendor will do it, guaranteed. Trying to get you to use their tools only, they'll fail to support the standard correctly. When that happens, we, as an industry, need to slap them.

Until then, we get to continue to flourish in Nirvana.

Don.

/reading: Programming Ruby - free electronic version.

Share this post :

Cash for Open Source? HuH?

Don MacVittie — Mon, 14 Jul 2008 22:15:43 GMT

There's a blog post - ironically touting a title with "economies" in it - over on C-NET.com talking about the need to get more enterprises to contribute cash and updates to Open Source Projects.

So the Freeks (Free Freaks who expect everyone else's work to be free but still want to be paid for their work) are now not-so-free? You can't have it both ways. Either capitalism in all of its glory reigns, or socialism in all of its glory reigns. But you can't destroy the free market and replace it with a mandatory payment scheme. Well you can try, if you're really interested in seeing your experiment dwindle on the vine. Just ask any of the (former) economists for the (former) Soviet Union.

Look, I use Open Source, I've contributed back to Open Source, but the point of OSS was to make software free. You did that. You start to sound like a drug dealer when you call for something in return for free software.

After all, the first hit is always free, right?

Don.

/Reading: This was a bonus post, see my last post from this morning if you actually follow my reading habits.

Share this post :

Storage - Where do we go from here?

Don MacVittie — Mon, 14 Jul 2008 17:10:57 GMT

There's a lot out there right now that is up in the air... The future direction of social networking, mobile vs laptops, security, when next I'll shave my head, and the topic of today's post, what to do about data storage.

There's a post over on DrunkenData.com that talks about preliminary results from a bit of UC/NetApp research that decided... SURPRISE! That 90% of your data is just sitting there taking up space.

The first thing that struck me about this claim was "Structured or Unstructured?" but I quickly remembered that it's NAS, so unstructured.

Not that this is really news to most people in the storage world, but the next question must be "what to do about that"? It's information we've had for a while, and things like ILM did nothing to help us capitalize on it, just as many of us predicted and analysts denied.

The post suggests going to tape and optical jukeboxes for those infrequently accessed files, and now that all OS's actually honor access times, that's possible in a NAS world (it wasn't, more recently than you think), and there are plenty of products like Acopia that can direct data based on access date/time. Another thing that gave me a chuckle was the tape world. Tape is good for long-term, off-site, archival storage. Its other uses have dwindled as the cost of disk has dwindled. But the article calls tape arrays "greener technology". Funny, it used to be sold as "cheaper". one can't help but wonder what it will be sold as after this post.

Another solution that doesn't require 15 year wait times for tape to seek to your users' files when they need access to them is disk arrays that spin down when not in use. Simple and to the point, EMC and NexSan make them, I'm certain others do too, but it's been a while since I was living/breathing in this space. The theory is simple, if this is tier two (or three) of your data architecture, and drawer #7 hasn't been accessed in X number of minutes, spin it down and leave it that way until someone does access a file stored there. That's green, it saves you green, and you don't have tape seek times to contend with (and in a jukebox, just tape or disk selection time could constitute the need for a coffee break).

Combine that with an intelligent tool for figuring out what goes on this system and what goes on your big, beefy, NetApp box, and you're in data nirvana, right? Well, at least your costs are down, your access times aren't up, and you are on the road to a greener Mother Earth. And tape can still be your archive format of choice.

I'm not as close as one might expect with the Acopia team (I know, I'm working on it), in fact, I know more about the competing technology one of our competitors sucked up, but migration based on access times is pretty simplistic as NAS Virtualization goes, so I'm going to stick my neck out with some amount of bravado and suggest we can do it.

So the architecture then would be simple. in front an Acopia box, behind it your spin-down disks on the right, and your NetApp Filers with hyper-active data on the left. As the data ages, it moves to the right (any presumed allegory to humanity is in the author's imagination, I assure you), the trays on the right are arranged such that all older data is in one tray, working backward to the newest stuff. So you'll have a couple of trays that never spin down, a bunch that are nearly always spun down, and your NetApp cranking out file after file. All coordinated by your NAS Virtualization engine, of course.

The only thing left in the seven-headed hydra that is data storage is SAN, and since the only company that truly managed the SAN/NAS leap is dead and gone, their technology and patents disappearing into the bowels of EMC, I guess you'll have to wait for the next revolution to work SAN into your NAS architecture.

Think about it, once set up you would have to do nothing and your electricity costs would go down. Monthly savings until such time as technology eliminates this architecture as an option versus the one-time expense of setting it up. Appealing.

Don.

/Reading: Smotherhood. These days I read while waiting for paint to dry on miniatures, and Lori left it lying around. Good book.

Share this post :

Trying out some new toys.

Don MacVittie — Wed, 09 Jul 2008 21:01:22 GMT

So Lori has a way to insert social media icons into her posts, I thought I'd try to one-up her. SO this Bookmarking widget for LiveWriter just... Does it. All the stuff at the bottom of this page is created by Insert Bookmarking Tags from the WLW Plug In directory.

I'm lazy, this was easy. I like.

I also finally downloaded the Polaroid blogging tool, here's a test of how it works...

Seems pretty cool to me. And yes, that's our youngest child in the picture.

So all in all, good stuff. Assuming these links work after I hit post...

Share this post :

Added the "share" button after the fact from: Add This That one is from Lori - thanks!

Don.

Technorati Tags: Windows Live Writer, Tools, Blogging Tools, dmacvittie

Online and Old Media Redux.

Don MacVittie — Tue, 08 Jul 2008 17:00:46 GMT

Heh. Good article about the state of publishing on the Associated Press. Seems someone has said all this before.

Here's hoping they figure it out, because they're not the only ones suffering from the problem.

Don.

Twitter Uptime: A Personal Solution.

Don MacVittie — Mon, 30 Jun 2008 19:54:20 GMT

I've been Tweeting more or less for the better part of a year.

Twitter has been down, more or less, for the better part of a year.

Clearly this organization did not take their rate of growth into account when designing (and redesigning) their architecture. No matter what they claim.

If I knew who our salesperson was for them, I'd send 'em off with a personal note, but even though I've met most of our salespeople at company functions, I only feel like I actually know two of them.

So instead, to help Twitter resolve their problems, I'm done Tweeting.

I'll take the month of July off, reduce their bandwidth, save them from the burden of carrying my oh-so-clever commentary and oh-so-important updates across their servers.

I suggest you do the same. Give them the opportunity to fix the problems, reduce their burden!

When you think about it, there's not much to Tweeting anyway. What can be said in 140 words is rarely enlightening or even impactful.

So if you're one of the people who follows me on Twitter, see you in a month.

If you're not, sign up and keep me honest. I sometimes find that I have something to say that I absolutely must Tweet to share with the world (though often the world doesn't care, that's the nature of Tweets!).

Don.

A long, long line.

Don MacVittie — Wed, 25 Jun 2008 20:37:13 GMT

"Hey, this line's pretty long, isn't it?"

"Yep, been backed up like this all day. Only a few of us are getting through at a time. It's pretty ugly."

(Looks at watch) "Wonder if I'll get there on time, I've got a deadline to meet"

"Don't count on it. Watch as those ahead leave the line and come back later. Only thing is, then they start over, and the line is getting longer, not shorter."

"Why are they carrying a body up there?"

"Have to keep its place in the line. It's a standard we all agreed upon, carry the expired along with us. No worries though, every one of them that passes its time means we'll get up there faster, because they don't have to get in, just have to be checked to make sure they're really expired."

(looks at watch) "Man, this is taking forever."

"I've been here most of my life, and I'll likely die in this line. What are you complaining about?"

"Wait, what's going on up there? It looks like they're splitting the line up!"

"Oh my you're right, they've added processing points, we might all get in!"

"Look, the line is being split into five!"

"Now I know we'll all get in. Thank goodness, I was truly worried I wouldn't make it."

"Look at everyone streaming out!"

"Yep, those are ones being sent back to let everyone know we got here okay. You didn't see them before because so few were actually surviving the backed up line."

"And there, on the horizon, what is that?"

"That big red ball? I dunno, let me see.... It says "F5" on it." (shrugs)

"Whatever it is, it appeared at the same time as the lines split. Well, I'm being called over to that short line by the guys with that same ball on their shirt."

"Good luck, hope you find your place!"

"I will now, I'm sure of it."

(turns and walks to the short line, passing people headed away, all going "Ack Ack Ack...")

Sorry, it's one of those weeks. Hopefully it helps someone understand load balancing, but if not, hopefully you got a laugh.

Don.

/imbibing: Mt. Dew

/listening: Hair bands

/reading: D&D 4E, of course

Advertising on the cheap - Off Topic Warning.

Don MacVittie — Mon, 23 Jun 2008 20:08:49 GMT

Since we were just talking about this - or rather I was pontificating about it - on last week's podcast, I thought I'd comment on TechCrunch's article about Advertising Shift.

The interesting thing about this article is how surprised most people who comment are by the percentage of revenue on the web versus "traditional media". Anyone who has ever looked into the medium as a revenue stream knows that there is little to no money in web advertising, and that greatly explains the difference in dollars expended.

An entire magazine can run from the advertising revenue generated by putting company's advertisements in front of qualified readers. Lots of them do, any that don't charge you subscription fees. The thing is that outside the techno-geek world, there are still a ton of very popular magazines, it is only in high-tech that the web has destroyed this medium and left the remnants of a couple of once-popular pubs limping along. Other areas of publishing have felt it, but like the breeze before the storm, not like the tsunami that swept through high tech publishing a year or two ago.

But in many respects, we broke the system. Magazines that provided useful services to readers went away, not out of lack of demand, but from lack of revenue. And you see, there is a problem. Web sites that ate the advertising budget that traditionally went to print publications sprang up on little or no budget, with the view that it didn't take much. And it doesn't until you are a professional publication and you need graphic artists and editors and other support items. The cost of a blog on a free hosting site is essentially just your time. But that's not a publication, it's a rambling of questionable quality.

And that's where the rub comes in. You see, since web pages that were start-ups could do everything on the cheap, they offered advertising on the cheap. A real life example (stripped of vendor name, but it was not F5) is that a major international tech firm paid $45,000 for a spread in a print magazine, and in the same month complained that $0.0075 per view of a qualified user - an IT person known to have an interest in their products - was too much. Let's ponder that for just an instant. In order to make up for that one single two-page advertisement, a web publication would have to generate 6,000,000 clicks. It doesn't take an IBM Senior Fellow (Hi Sach!) to figure out that it is highly unlikely for a web publication to generate the revenue that print publications did. And they will need to if they want to grow. You cannot continually pile more advertisements onto each page and increase your revenue, so you have to have more pages with content. And more content comes from man-hours. In high-tech, that content comes from relatively expensive man-hours.

The rub? Things tend toward pop-culture. Throw a few video game articles into your high-tech site, and they'll generate more users which equals more clicks - but it also means the users seeing vendor's advertisements are no longer qualified users, and a smart vendor will insist on paying even less, for it has been true forever that targeted/qualified advertising is worth a lot more than generic advertising. The kid looking for WoW hacks couldn't possibly care less about F5's products, using us as an example.

And it will have to change. Those who still wear their rose-colored glasses and toast the demise of traditional media miss the point - something is being taken from us, and something will have to replace it. And that something has to generate enough revenue to be profitable. What it will be, I don't know, but a new model will have to spring up, or online advertising rates will have to raise, because websites that appeal to everyone on the planet see a lot of traffic but are unlikely to offer specific information on your topic.

I was an employee of Network Computing Magazine for years and a freelancer for years before that, so I'll offer them up as an example. They used to be an independent publication with five test facilities costing hundreds of thousands of dollars to put together and tens of thousands each per year to maintain. Now they have a single lab and are folded into InformationWeek. If you read the labs pages of InformationWeek you'll see that they rarely have high-end hands-on testing anymore. Why? Because revenues dipped to the point that they couldn't maintain labs and staff capable of testing the latest 10Gig gear or the coolest new storage array. So now where do you go for unbiased test results? Tolly is paid by vendors to do their testing, Joe Bob Browning, guy who uses product X at work is not an unbiased source and normally can't provide hardcore test numbers. F5 ended up putting our testbed online so you could download it and do it yourself. Even if you have time for that, we're one vendor that effects maybe three subindustries with that test suite.

But web revenues (which from my admittedly limited understanding were not bad for the industry) did not make up for losses in print revenue. And so Network Computing went away. And you have less valuable information.

I'll be intrigued to see what business model springs up to replace seriously focused print magazines, but it's none of the ones out there today, and reading those numbers from Tech Crunch, remember that there's a lot more shift going on than dollars would represent, just because web advertising costs are a tiny percentage of traditional media, and in the long run that means people like you and I have less access to detailed and accurate information.

Don.

There's social networking, and then there's social networking

Don MacVittie — Wed, 18 Jun 2008 20:02:19 GMT

There is an interesting story over on InfoWorld about the valuation of LinkedIn. Funny how even tech publications can get confused and miss the point. It says that MySpace and FaceBook are "more conventional social media sites". Funny that, since FaceBook has been around for as long as LinkedIn, but MySpace hasn't been. All three were formed at about the same time. You'd think "traditional" would carry the flavor of tradition with it, but none of these companies is five years old yet. Traditional social networking would be BBS's.

More importantly, they miss the reason that LinkedIn will likely end up worth much much more than either of the other two - focus. Just as bulk mailings receive a much smaller response rate than targeted advertising, so is it true of websites. A site with a focus on the business person is granular enough for today (though not likely for the long term), while one that puts out a siren call to everyone will struggle to get people to respond. In short, I'd rather have access to LinkedIn's membership - which I joined voluntarily years ago - than MySpace - which I still don't have an account on - or FaceBook - which I joined this year because the boss said it was a good idea.

Simply put, the people on LinkedIn are business people trying to network. The other two mediums lack that focus completely. That makes the value of LinkedIn membership much higher to the average information worker.

My thought is that time will bear out LinkedIn's value to be significantly more than the other two sites this article compares to, simply because it is far easier to leverage a focused group. And over time I expect that even more fine-grained sites will become successful so that like-minded individuals can communicate effectively without being subjected to things like Gregor the 12 year old WoW player's rantings.

Of course, I'm an "old guy" in Internet terms - I still use email and get offended when my reviews website is called a blog - so what do I know? I know that DevCentral's strength is focus - that Gregor isn't interested in it, meaning we can all communicate about stuff that matters to us without interruption. So I guess I know something.

Don.

Good Stuff and Vacations.

Don MacVittie — Tue, 10 Jun 2008 00:19:03 GMT

Well, it's been an exciting week all around!

Amazon had their little stumble, and that got the conjecture/rumor mills running, Apple is talking about all sorts of new stuff, and the much bally-hooed 4th edition of Dungeons and Dragons was released (with a tribute to Gary, good call guys).

I read the news on Amazon and Apple, talked to some of my contacts to find out more inside information just because happenings in high-tech interest me, and stood in line at the FLGS at midnight with a guy lovingly called The Ninja to get our copies (and Lori's) of D&D 4E.

And personally, I got an article placed in a publication that I've thought I'd like to get something placed with for a while, on a topic that I'm passionate about, good stuff. Top all that off with the fact that I'm headed out on vacation as soon as I post this, and that's a pretty good week.

While I'm out, you'll need some reading - you do all come here just to read what I write, correct? So I thought I'd leave you some pointers to some of the other great stuff on DevCentral. Our team rocks, so finding great stuff isn't nearly as difficult as keeping it down to blog-post length!

There's another hit for the DC Post of the Week - If you haven't been following Deb and Colin on these, you should. They take a good thread from the forums and expound upon all of the possible solutions.

Colin has a great article on forcing download (as opposed to display) of certain file types: Content-Disposition. If I recall correctly, this article was spawned by last week's Post of the Week.

Lori brings your attention to a problem with some routers supposedly caused by Windows XP SP3 in her blog.

While we're speaking of Lori, I've probably mentioned this one before, but her Unifying Rest Access article is great if you're a BIG-IP user with a REST architecture.

And finally, Joe's bit on Self IPs gives you all you need to know about the Self IP interfaces for iControl. A good reference to have if you need to twiddle with self-IPs from code.

Me? I'll be at home catching up on all sorts of stuff I should have done months ago, then Lori and I are going to head out to Vegas for a couple of days to meet with some other geeks (we're bringing 4th Ed D&D books and our C&C books, so that kind of geeks).

Until then, enjoy the above reading, it's good stuff.

Don.

/reading: Keeper of Dreams, Orson Scott Card.