Thanks for the catch; but allow me to clarify:
Houston, Alpharetta (GA -- Atlanta), and Austin will be the 3 cities with redundant "lights out data centers" that will house HP's servers and storage for internal applications. In theory, only security personnel will be on site in these locations. Five of them are new; one will be a modernization of an existing facility in Houston at the former Compaq Campus.
The new facilities in Georgia are estimated to cost $240M.
HOWEVER, this is only for the internal applications.
Hosted applications outsourced by customers to HP Services today sit in 70 data centers in 23 countries. HP has not committed publicly (that I'm aware of) to a consolidation plan for these centers; however, it will be difficult for HP to remain competitive in that industry if they do not consolidate duplicate centers used for client applications.
I don't expect anyone to acquire HP. Compaq's acquisition of Digital created an enormous bloat that ultimately sank Compaq. HP's ill-advised decision to acquire Compaq was in the political and financial interests of a handful of executives at both firms, at the enormous expense of employees, stockholders, and customers. More significantly HP inherited the problems of a troubled Compaq in a troubled industry. Since the acquisition, HP's stock as risen as I predicted it would from $11/share to the low 30's. However, most of that rise as been "normal buoancy" of a rising tech market recovering from the 2000-2001 decline and the collective sigh of relief when Fiorina's hand-picked board of directors found balls enough to fire her.
Despite that progress (for which my retirement fund is grateful), HP faces daunting challenges:
1. Their market share in printers was so high that there was nowhere to go but down, as offerings from Epson, Canon, and others brought increasingly credible offerings to market.
2. The printer market itself has been saturated.
3. The PC business is only marginally profitable and unlikely to improve.
4. The large server market is (Unix SuperDome systems) is under pressure from increasingly powerful dual core offerings from Intel and AMS.
5. HP's multibillion dollar gamble on Itanium (remember HP partnered with Intel to co-invent and co-fund Itanium) has largely failed, as AMD forced Intel's had with it's dual mode 32/64-bit Opteron, leaving Itanium to join Betamax in the Hall of Fame for great technologies that the market passed on.
6. HP has huge customer credibility issues across an untenable array of platform and operating system offerings: multiple versions of Unix, Tandem Non-stop, DEC Alpha and it's myriad of also-ran OSes, and MPE, which has survived HP management's best efforts to kill it. It's not that customers don't understand the HP roadmap: it's that HP has earned low credibility.
7. Even if HP returned to its $11 five year low, the market cap is so large that only a stock swap in a highly inflated market would permit HP's acquisition. Even then, who could buy them without getting shot down by FTC or EEC antitrust regulators. IBM's big enough; Dell might be. But either would create untenable monopoly through an acquisition of that size. The only possibility of an acquisition I could forsee is from outside the IT Industry.
8. HP's profits still largely come from ink, toner, and print media -- an annuity revenue stream for HP, but one facing erosion as years of market share losses on print platforms translate into lower growth in ink.
I look for HP to begin selling off assets and lines of business.
On Telecommuting...
The folks HP is reeling back in are application developers, IT support, network management, etc., not the customer facing architects and field force. HP has realized, I suspect, that workspace costing formulas were the problem (for example, a 8x8 cubibcle in Houston "cost" the same as a 8x8 cube in Manhattan -- not exactly market reality). There are substantial costs involved with telecommuting (networking, local equipment that would normally be shared). More importantly, IT operations is a team sport that often requires pulling people into a room and hammering out an answer or an agreement -- much harder to do when employee's are changing diapers while on a con-call.
What's really driving this announcement is that HP is reducing the number of datacenters it operates from and unfathomable 87 to a still barely believable 25. If the telecommuting model were left in place, you'd have support people in one city theorhetically supporting a consolidated data center in another city. That just doesn't make sense.
In the years since Lew Platt left, HP has done some remarkably stupid things. However, this move isn't one of them. It's a necessary move to get both the internal and external cost structure in line with a very competitive IT Services business. The disparity
Slashdot Mirror
Thanks for the catch; but allow me to clarify: Houston, Alpharetta (GA -- Atlanta), and Austin will be the 3 cities with redundant "lights out data centers" that will house HP's servers and storage for internal applications. In theory, only security personnel will be on site in these locations. Five of them are new; one will be a modernization of an existing facility in Houston at the former Compaq Campus. The new facilities in Georgia are estimated to cost $240M. HOWEVER, this is only for the internal applications. Hosted applications outsourced by customers to HP Services today sit in 70 data centers in 23 countries. HP has not committed publicly (that I'm aware of) to a consolidation plan for these centers; however, it will be difficult for HP to remain competitive in that industry if they do not consolidate duplicate centers used for client applications.
I don't expect anyone to acquire HP. Compaq's acquisition of Digital created an enormous bloat that ultimately sank Compaq. HP's ill-advised decision to acquire Compaq was in the political and financial interests of a handful of executives at both firms, at the enormous expense of employees, stockholders, and customers. More significantly HP inherited the problems of a troubled Compaq in a troubled industry. Since the acquisition, HP's stock as risen as I predicted it would from $11/share to the low 30's. However, most of that rise as been "normal buoancy" of a rising tech market recovering from the 2000-2001 decline and the collective sigh of relief when Fiorina's hand-picked board of directors found balls enough to fire her.
Despite that progress (for which my retirement fund is grateful), HP faces daunting challenges:
1. Their market share in printers was so high that there was nowhere to go but down, as offerings from Epson, Canon, and others brought increasingly credible offerings to market.
2. The printer market itself has been saturated.
3. The PC business is only marginally profitable and unlikely to improve.
4. The large server market is (Unix SuperDome systems) is under pressure from increasingly powerful dual core offerings from Intel and AMS.
5. HP's multibillion dollar gamble on Itanium (remember HP partnered with Intel to co-invent and co-fund Itanium) has largely failed, as AMD forced Intel's had with it's dual mode 32/64-bit Opteron, leaving Itanium to join Betamax in the Hall of Fame for great technologies that the market passed on.
6. HP has huge customer credibility issues across an untenable array of platform and operating system offerings: multiple versions of Unix, Tandem Non-stop, DEC Alpha and it's myriad of also-ran OSes, and MPE, which has survived HP management's best efforts to kill it. It's not that customers don't understand the HP roadmap: it's that HP has earned low credibility.
7. Even if HP returned to its $11 five year low, the market cap is so large that only a stock swap in a highly inflated market would permit HP's acquisition. Even then, who could buy them without getting shot down by FTC or EEC antitrust regulators. IBM's big enough; Dell might be. But either would create untenable monopoly through an acquisition of that size. The only possibility of an acquisition I could forsee is from outside the IT Industry.
8. HP's profits still largely come from ink, toner, and print media -- an annuity revenue stream for HP, but one facing erosion as years of market share losses on print platforms translate into lower growth in ink.
I look for HP to begin selling off assets and lines of business.
On Telecommuting...
The folks HP is reeling back in are application developers, IT support, network management, etc., not the customer facing architects and field force. HP has realized, I suspect, that workspace costing formulas were the problem (for example, a 8x8 cubibcle in Houston "cost" the same as a 8x8 cube in Manhattan -- not exactly market reality). There are substantial costs involved with telecommuting (networking, local equipment that would normally be shared). More importantly, IT operations is a team sport that often requires pulling people into a room and hammering out an answer or an agreement -- much harder to do when employee's are changing diapers while on a con-call.
What's really driving this announcement is that HP is reducing the number of datacenters it operates from and unfathomable 87 to a still barely believable 25. If the telecommuting model were left in place, you'd have support people in one city theorhetically supporting a consolidated data center in another city. That just doesn't make sense.
In the years since Lew Platt left, HP has done some remarkably stupid things. However, this move isn't one of them. It's a necessary move to get both the internal and external cost structure in line with a very competitive IT Services business. The disparity