Breaking News
Comments
Newest First | Oldest First | Threaded View
<<   <   Page 4 / 8   >   >>
goafrit
User Rank
Manager
Re: leadership at HP
goafrit   1/4/2014 10:40:25 PM
NO RATINGS
>>  What would you like to see out of the company?

When Hurd was running the firm, it was expanding and growing. The fact is this, after the man left, all these issues came up. They went and bought one English balloney in Autonomy and ever since, they have not recoverred. If shame can be taken away, they need to bring the man back if he will agree.

junko.yoshida
User Rank
Blogger
Re: An incompetent CEO at work
junko.yoshida   1/4/2014 10:34:09 PM
NO RATINGS
@DMcCunney, as usual, thank you for posting such well argued perspective. (I relish every point you make here.)

I think you nailed it.

HP's case is more complex, starting with the question of just what HP is.


HP, indeed, has grown into just too many things to many people (internally and externally) without a clear focus. That, I think, is the crux of the issue.

DMcCunney
User Rank
CEO
Re: Happy new year
DMcCunney   1/4/2014 8:34:00 PM
NO RATINGS
@David Ashton:Let's clear up a little confusion.

Technically speaking, Linux is the Linux kernel, vmlinuz.  It it uses a Linux kernel, it's a Linux system.  My old Linksys WRT54G router was a Linux system, using a Linux 2.6 kernel. Because it was a Linux system, the firware was open source under the GPL,  people could grab and hack it, and did,  I ran a third party package called Tomato that would let me SSH to a command line in the router, and run a cut down version of th vi editor to diddle config files.

Current Linux distros will largely use the same kernel version, and the differences will be elsewhere.  All will have a Linux kernel, most will come with a set of the Gnu utilities, and they will differ in the window manager used (and therefor the GUI presented to the user) and the bundled applcations.

Linux distros intended for low end hardware like Puppy Linux or TinyCore will be optimized for size on disk, and will include utilties based on Busybox rather than the Gnu tools.  (Busybox is a set of cut down versions of the standard utilities packaged as one large executable, with individual utilities being symlinks calling Busybox with the appropriate parameters to exercise the particular functions.)  An example is Damn Small Linux, whose guiding principle was whatever could be fit in a 50MB ISO distribution file. It reached the point where hte developer simply couldn't pack anything more in.)

I recommend as much RAM as you can give the machine, with 512MB a good minimum and a gig or two being better.  As mentioned, the machine will be I/O bound.  Linux uses a memory paging system similar to Windows, where memory is allocated in 4K pages.  Linux memory consists of physical RAM plus the size of the allocated swap file on disk, similar to Windows virtual memory being installed RAM plus whatever size pagefile.sys is.  If there is more going on than will fit in physical RAM, Linux, like Windows, will swap less recent used memory pages to disk to make room.  If something tries to access a page no longer in memory, a page fault occurs, and the swapped out page(s) are swapped back in, with others swapped out. 

RAM is an order of magnitude faster than disk.  You want to minimize disk I/O, and that means reducing swapping and running a good sized disk cache.  The more RAM you have, the less need there is to swap, and the larger a cache can be maintained.

Your AGP graphics card ought to be supported, but I'm not sure about SCSI.

Get an ISO for an older Ubuntu release, like 11.10, burn it to CD, and see what happens when you try to install it. (The most recent Ubuntu release bit me when I tried to upgrade the Lifebook to it. It requied PAE support in the CPU, but installed everything else before it discovered it wasn't there and couldn't successfully upgrade the kernel.  That failure caused all manner of fun side effects.) If it works, you have a running Linux system.  If it doesn't, we can advise further

DMcCunney
User Rank
CEO
Re: An incompetent CEO at work
DMcCunney   1/4/2014 7:40:37 PM
NO RATINGS
@Junko: @Simon7382, you wrote:

Organizations can reach a point of decline when there is no return. Just think about Sun or Silicon Graphics, etc.

I agree. I do, however, wonder how organizations reach that point. I mean there are a lot of business management books talking about how to succeed in business, but how corporations die seems like just as important study as the birth of startups/corporations.


There are endless case studies of individual failures.  The problem is drawing generalized conclusions.

Sometimes the companies involved had good management, but were swept under by abrupt changes in the market.  An example might be Digital Equipment Corporation, once the second largest computer maker after IBM.  DEC had good management, all things considered.  (You will find folks who chart DEC's decline from CEO Ken Olsen's decision to kill the Jupiter project, intended to create the successor to DEC's 36 bit DEC-10 and DEC-20 machines, but I think that's misplaced.)

DEC was caught be the same thing that has caught many other computer outfits: the advance of technology, and the need by customers to do things cheaper. DEC's flagship product was the 32 bit VAX minicomputer, based on the proprietary DEC LSI-11 processor architecture.  But a new generation of "super micros", based on the Motorola 680X0 architecture and running flavors of Unix were appearing.  They could to the work VAXen did in many cases almost as well, and instead of spending $250K for a VAX running VMS, you could get a super-micro to do much the same thing for a fifth to a tenth of the price.  (PCs increased in power to cannibalize the super-micro market later, and brands like Apollo, Pyramid, and MassComp are footnotes in computer history.)

It was also the heyday of the RISC processor, with Sun's Sparc, HP's Precision Architecture, and IBM's Power line among others being developed.  DEC had its own well regarded entry, the Alpha.  The market was making a mass migration to cheaper hardware.  DEC had its own Alpha based workstation.  The problem was ramping up Alpha sales fast enough to stem the bleeding and replace the revenue form plumetting VAX sales.  DEC couldn't do it in time, (in part because Alpha boxes were a lot cheaper than VAXen and it had to sell a lot more to take up the slack).  DEC sold off pieces of itself to stay alive, and what was left was bought by HP becore if in turm was merged with Compaq.

There are also examples of "Should have seen it coming."  You covered the problems faced by Sony, Panasonic, and Sharp as the big screen TV market declined.  All three had made enormous capital investents in the plant neeeded to produce the big screens, and all three did well while the market was new and growing, people were replacing old TVs and trading up to new ones, and they could charge high prices and make high margins.  But like everything else in consumer electronics, the market became saturated, and the products became commodities, where the purchase decision came down to price.  Outfits like Samsung could address that market because they had lower costs.  Sony, Panasonic, and Sharp could not, and started hemoraghing.  The question is why the Japanese outfits didn't see this coming and have an exit strategy in place, since it would happen, but Sony, Panasonic, and Sharp are hardly the only ones that wear such blinders.  If things are going well, there's an almost inevitable tendency among managements to assume that state of affairs will continue, and overlook or ignore evidence of impending change.

HP's case is more complex, starting with the question of just what HP is.  Is it a supplier of equipment to business (the test and measurement business spun off as Agilent, or the high end server and printer lines)?  Or is it a consumer products company selling through retailers?  They are very different businesses with different models and sales and product cycles.  HP drifted along by inertia, trying to make incremental improvements in existing lines of business, but paid insufficient interest to indicators of change.

Leo Apotheker blamed HP's woes on the acquisition of Palm, but blaming the outgoing CEO for the problems is no surprise.  I think the underlying weakness was there prior to that acquisition, and would have bitten HP in any case.

HP needs to rethink who it is, what it does, and who it does it for, and I suspect a successfully revived HP might look a lot different than the present company. The only certainty I see is that HP can't continue as it is.


Simon7382
User Rank
Freelancer
Re: An incompetent CEO at work
Simon7382   1/4/2014 5:58:51 PM
NO RATINGS
Junko, that is a very interesting point. Maybe you or one of your colleagues should research the topic and write a book about it. One of the business books that I have read (and liked a lot) which has relevant observations on this topic is Andy Grove's "Only the paranoid survive". His theory about inflection points in businesses and that some companies come out of these rejuvenated and stronger while others miss it and die may be one of the explanations. For example DEC's history and demise seems to be a perfect illustration of Grove's inflection point theory.

seaEE
User Rank
CEO
What impressed me about HP
seaEE   1/4/2014 5:09:22 PM
NO RATINGS
What impressed me about HP was that they started out in test and measurement, but ultimately they were highly succesful in the calculator, PC, printer, and electronic component businesses (LEDs and communication IC's).  Somehow they were able expand upon those technologies that fed into their original core competency.  To do that takes a "go for it" mentality along with seasoned good judgment.  I think there is as well something of the "muse effect" that inspires engineering as well as the arts.  Einstein was noted for his long walks in the countryside during which he thought.  Beethoven was noted for his long walks in the countryside during which he composed.  Most time cards to not have a charge number for long walks in the countryside, and yet you need a certain amount of freedom to be able not just to think, not just to engineer, but to muse..."what could happen, what might happen, what if this leads to this".  Oops, somebody is coughing at me.  Back to work!

betajet
User Rank
CEO
The Law of Large Organizations
betajet   1/4/2014 9:40:54 AM
NO RATINGS
junko.yoshida asked: I do, however, wonder how organizations reach that point.

I think the primary reason is the Law of Large Organizations:
In any Large Organization, Loyalty will always be rewarded over Competence.

Employees who do not follow the Way We Do Things and warn of Icebergs Ahead get weeded out as trouble-makers.  Managers surround themselves with people who tell the managers What They Want To Hear, and then management acts surprised when the company starts hitting icebergs: "No one could have predicted that..."

JMO/YMMV

junko.yoshida
User Rank
Blogger
Re: An incompetent CEO at work
junko.yoshida   1/4/2014 6:51:47 AM
@Simon7382, you wrote:

Organizations can reach a point of decline when there is no return. Just think about Sun or Silicon Graphics, etc.

I agree. I do, however, wonder how organizations reach that point. I mean there are a lot of business management books talking about how to succeed in business, but how corporations die seems like just as important study as the birth of startups/corporations.

David Ashton
User Rank
Blogger
Re: Happy new year
David Ashton   1/4/2014 12:46:59 AM
NO RATINGS
@DMcCunney...thanks again....I am looking at a desktop, probably around 1GHz, I could probably get 500MB to 1GB RAM, a decent PCI or AGP graphics card.     As an example I have a machine I used to have Win 2000 on, it's got a Gigabyte GA6BX motherboard,  an 800MHz P4 processor, 768MB ram and an AGP graphics card (not sure which one without firing the machine up).  It's got a SCSI card in it and I used to run a SCSI 12GB tape drive for backup,  If that (as a machine) would work I'd use it.  Would that run Ubuntu or should I look at something smaller (Linux-wise)?

I'd probably not use Multiboot, I have a KVM switch so could switch between machines fairly easily.... 

DMcCunney
User Rank
CEO
Re: Happy new year
DMcCunney   1/4/2014 12:29:06 AM
NO RATINGS
@David Ashton: I have in mind around a  1-GHz machine (which would be hopeless by windows standards) but many Linux distros appear to want more than this? 

The scarce resource will be RAM, not CPU speed.

I run Ubuntu on an ancient (in computer terms) notebook: A Fujitsu Lifebook p2110 circa 2005.  (It was a pass along from a friend who upgraded but didn't want to just throw it out.)

The Lifebook has an 867mhz Transmeta Crusoe processor, ATI Rage Mobility graphics with 8MB video RAM, a 40GB IDE4 HD, and a whopping 256MB of RAM, of which the Crusoe grabs 16MB off the top for "code morphing".

I reformatted the HD, and multiboot Win2K Pro, Ubuntu, Puppy Linux, and FreeDOS.  Ubuntu and Puppy are on ext4 slices and mount each others partitions on boot, so I have the ability to install some apps once but access them from either OS. Both can see and access the NTFS slice Win2K is on

Win2K is on NTFS, but I found an open source driver that lets me read and write to the ext4 partitions from Windows.  (Win2K actually runs acceptably in 256MB RAM.  WinXP SP2, one the machine when I got it, does not.)

All of them can see the FAT32 slice FreeDOS lives on.  FreeDOS can't see anything else, but I don't care.

Getting Ubuntu running on the machine was a challenge.  What I wound up doing was installing from the Minimal CD to get a working CLI installation, then using the apt-get package manager to pick and choose the rest.  I use LXDE as the desktop, since it's the lightest weight one available.  Ubuntu package management is a strong point: select a package, and it looks at your system for dependencies, and automatically includes eerything the package needs to run.  Selecting LXDE brought the needed X-Windows environment with it.

The issues I run into revolve around low RAM and slow HD.  Linux likes lots of RAM.  Ubuntu runs in a low RAM system, but has to do a lot of swapping.  (The Lifebook can be upgraded to 384MB, but I could buy many gigabytes of RAM for a more modern system for what the 128MB upgrade touwld cost.)   The IDE4 HD is a BIOS limitation.  Swapping in a faster drive isn't an option.  Performance of small apps is acceptable.  large ones are problematic because of slow HD access.  (I don't even try to run a current Firefox build - it takes 45 seconds to load and initialize, and is perceptibly slow when up.)

The machine is purely a toy to see what performance I can wring out of ancient HW without throwing money at it, so I'm not unhappy.  I had low expectations going in.

If you have a gig or so of RAM and a reasonable HD, Linux shouldn't have a problem.  Your likely issue will be video.  Ubuntu Unity, for example, wants a graphics card with 3D acceleration.  It won't run on the Lifebook.  It does work on the desktop with integrated Intel graphics, but I don't care for it.  An interface intended to make best use of small screen sizes on notebooks and tablets falls down on big monitors.

Unless you are doing demanding processor intensive stuff, CPU speed is less critical than you might assume.  Most desktop machines are I/O bound, not compute bound , and the CPU spends most time in a wait waiting for I/O to occur.  All machines wait at the same speed. :-)

Tell me more about the specs of the machine you plan to put Linux on and I can give better advice.


 

<<   <   Page 4 / 8   >   >>


Flash Poll
EE Life
Frankenstein's Fix, Teardowns, Sideshows, Design Contests, Reader Content & More
Rishabh N. Mahajani, High School Senior and Future Engineer

Future Engineers: Don’t 'Trip Up' on Your College Road Trip
Rishabh N. Mahajani, High School Senior and Future Engineer
Post a comment
A future engineer shares his impressions of a recent tour of top schools and offers advice on making the most of the time-honored tradition of the college road trip.

Max Maxfield

Juggling a Cornucopia of Projects
Max Maxfield
2 comments
I feel like I'm juggling a lot of hobby projects at the moment. The problem is that I can't juggle. Actually, that's not strictly true -- I can juggle ten fine china dinner plates, but ...

Larry Desjardin

Engineers Should Study Finance: 5 Reasons Why
Larry Desjardin
28 comments
I'm a big proponent of engineers learning financial basics. Why? Because engineers are making decisions all the time, in multiple ways. Having a good financial understanding guides these ...

Karen Field

July Cartoon Caption Contest: Let's Talk Some Trash
Karen Field
127 comments
Steve Jobs allegedly got his start by dumpster diving with the Computer Club at Homestead High in the early 1970s.

Top Comments of the Week
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)