Capitalizing SaaS

Filed under: Random Thoughts — barmijo — June 29, 2007 @ 12:50 pm

There are two types of SaaS companies, those that are really ISVs trying to move to the Web and those that are really Web 2.0 companies trying to tap what’s traditionaly been the ISV market. Ironic perhaps, but true.

Web companies calling themselves SaaS are easy to spot – they have no ISV legacy, no perpetual license customers. IMHO, they appear to classify themselves as SaaS as part of attracting higher-end business clients. What separates these companies from other web concerns are their need to deal with customer issue related to data security, business continuity, off site backups and the myriad of other enterprise IT issues that they inherit from their target clientelle. Tough issues, true, but surmountable.

ISV’s trying to move to the Web, however, face a what until now has been a potentially larger hurdle. As established businesses they have fixed expenses and depend upon a steady revenue stream from licenses to cover their cash needs. Converting to subscriptions potentially introduces a chasm in their cashflow. Navigating that chasm must look to a CEO like jumping the Grand Canyon on a tricyle.

That may be changing, though, as I recently got a google alert (love that tool) that lead me to SaaS Capital, who finance SaaS receivables. This is the first firm I’ve seen specifically targeting this space with a greatly needed service. So if you find yourself on that tricyle and hurdling down the ramp at breakneck speed with the chasm ahead of you – they’re worth a look.

Another post on startups from Marc Andreesen

Filed under: Random Thoughts — barmijo — June 27, 2007 @ 11:20 am

Another great post in Marc’s series in which he makes the point:

The only thing that matters is getting to product/market fit. . . . being in a good market with a product that can satisfy that market.

And you can always feel when it’s happening. The customers are buying the product just as fast as you can make it — or usage is growing just as fast as you can add more servers. Money from customers is piling up in your company checking account. You’re hiring sales and customer support staff as fast as you can. Reporters are calling because they’ve heard about your hot new thing and they want to talk to you about it . . .

The iPod, IMHO, exemplifies PMF. It wasn’t the first, and perhaps wasn’t the best technically, but clearly Apple read the market better than anyone.

Having gone through this ringer a few times, though, I’m often concerned by how people interpret having a good market. Many folks fall into the trap of waiting to see customers asking for product XYZ. While this may work for products that take a couple months to build, the market may shift completely before you can get a more complex product built.

Guardians of the English language

Filed under: Random Thoughts — barmijo — June 25, 2007 @ 1:20 pm

No one has ever accused me of being a stickler for proper vocabulary and grammar. Still, I read the news with
interest last year when Webster’s added google to the dictionary because my friends and I use google to settle disputes about spelling by searching for both alternatives and selecting the most popular. Ah, grammar by the crowd. My elementary school teachers would be horrified!

Recently, this trend has moved from the tech geek crowd into the mainstream. My oldest son now uses google as a reference tool as well, as are his friends. So it seems google not only succeeded in making it into the dictionary, it’s now actually replacing it.

What I didn’t understand until I became an advertiser, though, is that Google recognizes this power and is actively using it.
Spelling, punctuation and grammar are all enforced in Google ads. While this seems reasonable, and most of the time we appreciate being spared garish ads, at the same time it results in the Googlecrats creating an enviable position for themselves. The power to restrict the use of new words in our our tech-crazy what’s-new-today world can limit the reach of ideas and products.

A seemingly benign case in point is 3tera’s use of the word “datacenter.” Search and you’ll see that it’s in use by Microsoft, IBM, Sun, Oracle, VMware, and of course 3tera. It’s also commonly used by trade press and analysts. Google, however, doesn’t seem to recognize the word and placed one of our ads on hold with an error message that I haven’t used spaces correctly. Evidently “datacenter” isn’t in the Google dictionary.

The inconsistancy is truly ironic IMHO. Google has become the defacto method for researching anything. Yet, they don’t seem to trust their own results. If they did, the millions of uses of the term datacenter on sites with high page ranks should be sufficient to allow advertisers to use the term in our ads. But, alas, that’s not the case.

Why you shouldn’t found a startup

Filed under: Random Thoughts — barmijo — June 19, 2007 @ 1:21 pm

Marc Andreesen’s written an excellent post about why you shouldn’t found a startup. It should be required reading for anyone considering creating or joining a startup. Those of us silly enough to do this again and again should have it tattooed in reverse on our foreheads!

However, Marc forgot my favorite drawback. Startups are addictive. Their a drug. And, if you’re genetically predisposed, like me, once you’ve tasted startup life there’s no going back.

Each time I start or join a new fledgling tech company, it’s almost as if nobody knows me anymore. Suddenly I’ve got no money to spend and email responses take weeks, if they come at all. The idea, if it’s any good, is outlandish and unheard of (In my case Rapid City : routing in silicon, Topspin : datacenter virtualization switch, 3tera : utility computing in a browser). Folks just aren’t ready to abandon the familiar and accept your ideas. They need convincing evidence, but all you’ve got is an idea. Everything has to be created from scratch, from names and logos, to slideware and datasheets, and all by you. There’s no product, no money and no employees. So you incorporate, create your slide deck, seek a few other startup addicts and spend months touring conference rooms in silicon valley explaining to anyone that’ll see you just what it is you’re doing. It’s a loooooooong process.

One day, though, you’re introducing yourself at a conference when someone says “I’ve heard of you.” That’s a moment that knocks your socks off. You’ve created a little bit of buzz. Next you get the thrill of the first bits passing through your system. Seeing your product on the shelf at a major retailer, or on the screen of a stranger’s laptop for the first time is electrifying. Walking into a datacenter and seeing a hundred of your systems is indescribable. Having a writer print (or post) a review declaring that your widget defines the future is a rush.

Of course, we can’t forget the potential monetary rewards. I’ve been involved in two succesful acquisitions, and though the cars I bought are long gone, the smell of cigars still seems fresh from the post acquisition parties. And I’ll never forget the look on the tellers face when depositing the check (yes, I did get her number).

Knowing all the obstacles you overcome to achieve these successes changes you. The words no and impossible lose their meaning.

So, why is this a negative. Because, as I mentioned before, it’s addictive. Knowing what you can accomplish, especially with little money and a small team, makes working inside a large company incredibly painful. Endless approval cycles, countless reviews, competing initiatives, legacy products to consider . . . and all you want to do is produce something. Anything.

Although I can’t find reference to it online, Ross Perot, who ended up on GMs board after they bought EDS, was reputed to have blown up during a board meeting yelling “pick anything and be the best at it, even if it’s the ashtray.” Shortly thereafter he left to found Perot Systems. Likewise, most of my friends who’ve started succesful companies and been acquired last no more than a year before branching out on their own again – looking for another fix.

1TB RAM, 400CPUs, 40TB storage – 1 CLI

Filed under: Random Thoughts — barmijo — June 18, 2007 @ 4:50 pm

Enough people have commented on the scale of our new demo grid that I cobbled together a video feed via
so folks could can look in on our progress as we work to load it up. I’ve embedded the feed at right, but you can watch it directly or even embed it on your own page, if you’re really into this type of thing, with the following code:

<object classid="clsid:d27cdb6e-ae6d-11cf-96b8-444553540000" width="320" height="261" codebase=",0,40,0"><param name="src" value=",xNTxrpduaCaSxkyj.oSg.usc" /><param name="wmode" value="transparent" /><embed type="application/x-shockwave-flash" width="320" height="261" src=",xNTxrpduaCaSxkyj.oSg.usc" wmode="transparent"></embed></object>

SaaS and virtual Appliances

Filed under: Random Thoughts — barmijo — June 17, 2007 @ 11:57 pm

Phil Waineright recently posted on the use of virtual appliances as an enabler for SaaS to get inside the firewall:

“Delivering software as an appliance brings many of the same benefits as delivering software as a service. . . So, there’s a sense in which appliances are not so much a part of the SaaS model as competition for it.”

Rich Miller` commented further

“The bottom line is that SaaS has much to gain from embracing packaged software appliances, as both an infrastructure for on-demand service and as an on-site adjunct to their network-resident services.”

Folks are already using AppLogic to enable SaaS offerings without building the traditional multi-tenant application, so I read both posts with interest. I understand the desire to get “inside the firewall,” as Phil describes, but I’m concerned that he’s over simplifying when he says that vendors deploying virtual appliances on-premise “can push out a release to the entire network all in a single operation.” Major customers granting such ready access through their firewall for a vendor is outside my experience. We do, however, have users who’ve set up a virtual data center for a key customer as a secure operating environement.

If you’re operating a SaaS provider both posts are worth reading.

Facebook creates new market for utility computing

Filed under: Random Thoughts — barmijo — June 14, 2007 @ 4:26 pm

Like many folks, I’ve been following the initial experiences with Facebook’s API; their first salvo in what appears to be a coming platform war.
From all appearances, Facebook has done a great job with the launch, but may have exposed to many to the seldom discussed limitations
of the infrastructure that powers internet applications. Marc Andreessen has an extensive analysis of the impact that pretty much hits it:

“… code that uses the Facebook APIs has to run on third-party servers — servers that you, as the developer provide. …”

“… when your application takes off on Facebook, you are very happy because you have lots of users, and you are very sad because your servers blow up. ”

“… unless you already have, or are prepared to quickly procure, a 100-500+ server infrastructure and everything associated with it — networking gear, storage gear, ISP interconnetions, monitoring systems, firewalls, load balancers, provisioning systems, etc. — and a killer operations team, launching a successful Facebook application may well be a self-defeating proposition.”

Marc’s goes on to talk about ILike, the first hit app on Facebook. ILike had to add more than 100 servers in the first week after their Facebook launch. Just a few weeks later, ILike now has more than 6 million users!

Of course, 3tera’s been talking about (and building for) this for two years. Had ILike been a utility computing user, the now infamous post begging for servers would never have happened.

Defending services in the cloud

Filed under: Random Thoughts — barmijo — June 12, 2007 @ 10:32 am

With every advance in technology the same group of copyright owners show up to prove that they don’t understand or
care about their customers by trying to block technology that ultimately makes them more money. VCRs, DAT, MP3, TiVO, P2P all were
greeted with lawsuits. DAT was effectively killed by a combination of the threat of legal action, the DAT-Tax and, of course, the recordable CD. Napster was neutered. Now the litigous folks in Hollywood have turned their attentions to Cablevisions network based DVR service.
If you haven’t heard of this threat to innovation, you can start your reading with a recent eff article.

Cablevision’s service allows you to have all the functions of a DVR without the requirement of the HDD enabled set top box. Instead, the programs you choose are recorded by a device in Cablevision’s network for you. This is a great service for a few reasons. It’s lower cost, programs can become portable, and frankly who needs yet another box.

Not everyone is a television junky, so you may not feel the need for a DVR in the cloud, but this type of restrictive legal action threatens the innovation in our industry. Please check it out and make your opinion known.

The documented life

Filed under: Random Thoughts — barmijo — June 11, 2007 @ 2:40 pm

For some, blogging comes easy. My friend Isabel Wang, for instance, is adroit at posting well thought out and well written posts routinely. I, on the other hand, can go days between thoughts that I feel are worth cluttering the internet with.

And yet, blogging is still relatively slow and deliberate compared to new services like twitter.
If, like me, you find your self simultaneously both bewildered and fascinated by twitter, you might enjoy Nicholas Carr’s latest piece.

Another little demo grid – take 2

Filed under: Random Thoughts — barmijo — @ 10:50 am

We hit a couple snags last week with nodes that weren’t configured properly in manufacturing, but our latest AppLogic demo grid is just about complete. A lot of folks that we talk to have asked how the system performs as compared to a tradtional installation with physical devices – especially SAN. So we’ll be using this grid for running benchmarks that will be published around the end of the month. Along the way, though, I just had to share the numbers:


Older Posts »
This blog is powered by WordPress running on AppLogic standard LAMP cluster.   RSS feed