Tom Petrocelli's take on technology. Tom is the author of the book "Data Protection and Information Lifecycle Management" and a natural technology curmudgeon. This blog represents only my own views and not those of my employer, Enterprise Strategy Group. Frankly, mine are more amusing.
Wednesday, July 06, 2011
Friday, May 27, 2011
Friday, March 11, 2011
- Emerging – what every startup is. They don’t know what they will be when they grow up but that’s okay. This is where our most exciting technology comes from. Eventually, however, they will have to grow up and become something else or getting eaten but someone else.
- Parts Supplier – you make parts for other people. Like headlights or NICs. Great work, especially if you spread it out amongst a lot of companies. The goal of a Xyratex. Qlogic, or an Atto for that matter is to be a supplier to as many people as possible and build something of an aftermarket for your components. Think Cummings (they make engines).
- R&D Shop – companies that produce only intellectual property. You see this in Pharma and semiconductors but most computer tech companies want to control their R&D. If a big company sees something it likes in another company, they just buy it.
- Outsourced Services – who doesn’t outsource call centers and manufacturing these days? Most of the services industry falls into this category. It’s the business process equivalent of a parts supplier.
- Specialty Supplier – the big companies can’t make everything. High performance or special purpose products can’t be produced in enough volume for the big companies to be interested. We used to have more of these in the hardware industry. SGI was one and sort of still is. Alienware certainly was but was bought. Software is rife with specialty companies. Software can get away with it because they have almost no recurring costs. It’s all R&D and no inventory. What is important is that these companies have something that is very important to a small number of people but enough people to sustain the company. They are unique but have demand.
- Conglomerate – a set of loosely related companies. Some are completely unrelated like GE (aircraft engines and broadcast media?). EMC looks more like a conglomerate than anything else. You can try and put a “Data Management” wrapper around them but RSA, EMC storage, Documentum, and VmWare are only loosely connected in the marketplace. Conglomerates manage companies or divisions like a portfolio. They diversify to guard against downturns. Storage is down? That’s okay because security is up and so on. Google is looking more like a conglomerate every day. And when the need growth they buy some other company in a different space than where they are now.
- Solution Supplier – soup to nuts in a particular market. HP can provide you everything from mobile devices to laptops to storage to servers. Oracle and IBM can also provide you with almost a whole solution. For these companies, it’s a matter of defining your boundaries. Is is business hardware to business software like Oracle? Maybe it’s all software from infrastructure to desktops to game systems like Microsoft. It’s about delivering a complete end-user solution. When you need more growth, you push out the boundaries.
Monday, February 14, 2011
- Apple iOS – keeps going. Apple simply doesn’t care about the wider market. That and the cognoscenti love their Apple crack.
- Android – generous licensing will insure that it continues to evolve. It lives!
- Windows 7 Mobile – another failed attempt. Sorry Microsoft. I actually like Vista and Windows 7 on the desktop. The mobile OS is too little too late. It dies. Microsoft money insures it dies slowly and painfully. Please Mt. Ballmer, do a deal with Google and move to Android while you still can.
- WebOS – really? I get that HP paid good money for Palm but with all the other choices, why would I want this. The tablet market? And this from a company who’s last homegrown OS was HP/UX. It dies and HP switches to Android merging whatever is good about WebOS into it.
- Blackberry OS – this is a tough one. RIM has an enormous and fanatical installed base but it’s slipping. They had the first viable smartphone-like device which helped get them established. At the time you had to rely on their closed system for email. Now, that’s a liability. I’ll bet that they quietly move to something else but with Blackberry extensions so the old guard can feel happy. My guess is that it will be Android too.
Wednesday, February 09, 2011
Friday, January 21, 2011
ARM processors are subversive. They have been used in small, low power devices forever. They lie under the radar of general purpose computing. ARM is found in the stuff that you don’t think of as computers such as mobile phones, network switches, game consoles, and GPS systems. These little devils are the processor of choice for smartphones everywhere. Android was designed for it. It lives in the Apple iPad. There might even be one in your Blu Ray video player. ARM is everywhere.
It’s only a matter of time before ARM sneaks into the Empire of Intel, snatching away bits of key real estate. We saw that with the first Netbooks. Why? How could this happen? Power! It’s all about power. ARM cores provide that almost mystical combination of processing power and low energy use. Just look at the iPad. Inside is an Apple A4 based on four ARM cores. It gives the device all the power it needs while allowing it to claim a battery life of 10 hours while using WI-FI. Sure, that longevity is not only because of the ARM processor but it certainly is a key factor.
None of this would matter if it weren’t for the fact that mobile devices are taking over. I don’t expect servers or desktops or even laptops to go away anytime soon. There are times you need more raw computing power. But much of the time you don’t. For most home uses, a low-end, Internet enabled device that moves around with you is just fine. Truth be told, a lot of business users would gladly ditch the expensive and heavy laptop in favor of something small and light but long on battery life. The movement to cloud computing makes this even more appealing.
So where does this leave Intel? In mortal danger. Intel finds themselves, for the first time in ages, with an entrenched enemy on it’s doorstep. They are in the position of having to displace a competitor instead of having someone trying to eat their scraps. In the past, they could rely on revenue in all high growth segments, be they servers, desktop, or laptops. Not now. ARM has infiltrated core Intel markets as mobile computing devices supplant laptops and low-end computers.
For those of us used to the seismic shifts in the computer industry, the ARM ascendance is a bit surprising. We notice the Googles and Microsofts, the Oracles and Apples more. Companies that burst on the scene with all the subtlety of a Mongol invasion. The fifth column tactics of the ARM processor was not as noticeable. That’s why it’s so dangerous to Intel. ARM may have taken their market from within, quietly. It’s much harder to fight a threat that sneaks up on you.
Intel will certainly fight back. They have the technology, size, and money to fight a protracted war. The strategy has to be different from what they are used to. They are the underdog now. Given that, Intel needs to be more aggressive and less complacent. My advice to Intel is “shock and awe”. Buy and build whatever is necessary to crush ARM before it is too late. Or, one morning, all the residents of the Empire of Intel will awake to find ARM processors everywhere and themselves friendless and forlorn.
The enemy is at the gates! To arms! Oh wait. Wrong rallying cry.