Give us our beer back, HP! Not all, just a bit!

Although the title of your article was enticing the body of it was a poor attempt at humo(u)r to those of us feeling the morale problem in HP.

I'm sure you have heard from many of us on how to fix the problem. Beside the obvious answer on how to fix morale problem at HP here are a few more ideas.

Try and come within the ballpark when giving out bonuses to employees vs bonuses to execs.

I mean Carly gets a 200% bonus while the people who actually made the profits get 1.3%, and that was considered a "discretionary" bonus by the BoD since we didn't actually meet the goals set by the execs (yet they still got their full over-inflated bonus).

Give us our Beer back

I don't mean like in the early '80's when beer busts were a weekly event. Just a few suds 3-4 times a year to say "good job".

Quit micro-managing us

Before the merger, we as employees were pretty much left to do our job. We knew what our job was and in most cases our manager was clueless. This has been true for all time I believe.

Quit making us take Standards of Business Conduct, Total Customer Experience, Standards of Personal Conduct, etc, etc etc..

This is just busy work so someone in Human Resources can have a job. The content of the web based propaganda is common sense and a waste of our time. We do have a job to do and since many of our co-workers are now ex-co-workers, we are over loaded and over stressed.

These are just a few ideas. I could come up with more but let's start small and work our way up.

We just learned that Carly put HP's name back in the candidates list for Best 100 Companies to work for. This after being off the list since the merger. I should give her credit for doing this but with only a 1.3% bonus and no raises again this year, I'm the one that needs credit.

HP Guy

Check Point

Wow,

CheckPoint on the Inquirer, that's got to be a first.

In addition to CheckPoint FireWall-1 NG AI R55, for which you linked a hotfix, previous versions are also vulnerable. FireWall-1 NG FP3 is fixed by "HFA-323" (get it from your CheckPoint supplier or from CheckPoint tech support), while a fix for FireWall-1 NG AI R54 is still outstanding.

Users of R54 can wait until a fix is available, likely in HFA-407, or work around the issue: If H.323 is not used, block it completely at the firewall. If H.323 is used, restrict it to trusted hosts by Source and Destination addresses, and remove the H.323 service from triggering on the "Any" 'service'.

Cheers
Thorsten

Shorely Shome Mishtake?

To: Kieren McCarthy
Re: New Explorer hole could be devastating

Please obtain a dictionary.

You do not "reek havoc" you "wreak havoc". Reek means odor. I do not think the MyDoom worm is smelling havoc across the world.

If you use words for a living, at least use them correctly. Set a good example.

Otherwise, keep up the good work and I'll stop nitpicking.

Boris

Texas

I find your statement about living in Texas very ignorant. I use to enjoy your reporting, but now I can't believe any of your stories.

Richard Leonard

SPEC benchmarks

Hello Mike,

Funny it is mentioned that Cache might have a very high impact on the SPEC benchmarks. Have you ever had a good look at the spec benchmarks for the itanium? Although the differences in clock are not that dramatic, the spec numbers are... The higher the cache, the more absurd the numbers... I find the difference between the 1.5MB and the 6MB Cache versions rather stunning!

Greetings, Michel

China and Human Rights

Hi there,

I like your article "China throws more people into jail for online opinions" because this is the only way to keep the spotlight on the wrongdoes until their time comes.

But your article about Chinese ID cards may be a little bit biased. Before we evaluate what other governments are doing I think we should first verify that our own governments are clean.

In many European countries people are forced to keep an ID card ready. Now they may not be "chipped up" yet, but IMHO this is only a question of time. However, they all are machine readable. Whether your friendly officer retrieves a number and more from your ID card with chip or an ID card just with number which then points to a database really makes no difference.

All our travelling is being monitored. The European governments have just proven how much they care about our freedom.

The next step will be passports with RFID tags. There are already proposals for this approach being discussed between the US government and the European governments.

There is more. We _all_ are going to be fingerprinted and processed like criminals, not just the people who enter the USA. Say thanks the freedom loving US and European governments. Other countries are already retaliating and right they are, whether an Uebermensch of a US airline accepts it or not. The majority of people still thinks this won't affect them, maybe because they are not traveling. They are wrong. In particular US citizens have no clue what is coming. And neither do most Europeans.

The danger of this development is that we all are being monitored, not only the people in China. And every time when we have to provide an ID card or fingerprint we all have to prove our innocence. This is the perversion of justice. Of course there is always a thread at hand to justify everything. This is true now, and it has been 70 years ago. And never has the previous status quo been reestablished. How we are becoming numbers really does not matter, be it through a number on an ID card, a number derived from a fingerprint, facial imprint, iris scan or a tatoo on the arm.

Now, if we accept for a moment the need of ID cards, including passports, then the chip based solution may actually offer a solution to replace the number based monitoring. The point is that the ID cards only make sense if they are forgery-proof. A smart chip certainly can do this for a certain time period, at least much better than any other solution. And there would be no need to store the information in the hands of the Chinese government or the guys who currently occupy the White House.

There is really enough information on the web. You could easily add a few links to your reports at The Inquirer. I think it is important to report the issues, but there is no reason to focus only on the Chinese governments. Of course your state controlled television program will only show that "aliens" are now being fingerprinted to make us all feel save. Sleep well.

Henry G. Juengst

Via and Operating Systems

Greetings!

It's not really like the VIA CPUs don't support Comp-Exchange-8. They do. However, they can be configured (by BIOS or afterwards) to hide that capability from their CPUID feature map. This has been recommended by VIA and done by most BIOSes because Windows NT4 did (and possibly still does) not work when it sees a non-Intel CPU reporting cx8 capability, for whatever reason.

So all Linux actually would need to do is recognize the VIA CPU - it already does that anyway - and either assume cx8 is there, or tweak the internal configuration register to unhide cx8 capability.

It's all in the CPU datasheet, and it's been right there even back when the VIA C3 still was the IDT WinChip.

Quote:

Notes On CPUID Feature Flags
1. The CMPXCHG8B instruction is provided and always enabled, however, it appears disabled in the corresponding CPUID function bit 0 to avoid a bug in an early version of Windows NT. However, this default can be changed via a bit in the FCR MSR.

Peter M.

Thirty Stage Pipelines

Hmm - 30-stage (+/-) pipeline vs. 20, doubled L1 data cache, doubled L2 cache, SSE3, ALU will/won't (strike out where not applicable) run at 2x clock (It's not easy running a 16K L1 with a 2X clock.), .... Sounds to me like Prescott is really a Sextium 5 rather than 'the next Pentium 4'.

At the same clock, Prescott should be slower than Northwood, expecially if the rumor of the reduced speed ALUs is true. If Intel is introducing Prescott at the same clock as Northwood, they're really having trouble with the heat! [The P4 was not intro'd at 1GHz ! Who knows - maybe the reason Intle's calling the Prescott a 'P4' is that they'd look pretty damned stupid calling it a "Septium 5", and then introducing it at a lower clockspeed (and lower performance) than their top-clocked Northwood.]

The heat problem. Looking at the specs for max Amps and nominal voltage, it's obvious that under high loads, P4s already switch off and on to control heat generation. Prescott seems to be even worse (an even higher value for max Amps, and more stringent power supply specs), so when running a demanding app, it may only be running 70% of the time (and be 'off' 30% of the time) at its advertised clock while the already hot but relatively cooler Northwood might be able to run 80% of the time and be off 20% of the time at its advertised clock. (Actually, I think the cooling 'steps' are in 12.5%, rather than 10%, increments - in which case the 'duty cycle' comparo might be 62.5% vs 75%.)

Second problem - hyped threading, like single-threading, will also run afoul of longer stalls on a chip with a longer pipeline.

The large number of oddball fetches - and resulting cache misses - engendered by traditional apps are the reason that Athlons and PIIIs have always been better than the P4s at traditional business apps. (The PIIIs/Athlons are also better at humongous number-crunchers like Photoshop). For business apps, the Athlon/PIII advantage became less apparent when the 'benchmarks' for business apps were revised to maximize the streaming data processes that are present, but mostly little used, in those apps. [Not that I'm cynical (Yes, I am), but somehow I suspect that various 'business app' benchmarks will become even more loaded with streaming processes, in the very near future. [Photoshop is no longer commonly given as a single 'benchmark', but is mostly run as part of a video benchmark, and is not a big factor in the benchmark score.]

[I've seen some speculation on the wonderfulness of the idea of dual-cored, hyped-threaded chips, and I'm certainly glad I'm not faced with the task of developing the decision/control matrix for that can of worms!] :)

John

Windows Congregation should be banned
from House of Jobs

There are some from within the Windows congregation who should not be welcomed into the house of Jobs on account of their gross stupidity and unsanitary nature. These accursed folk must be regarded as being beyond salvation. Thou must have consideration for the the current denizens, who have fashioned unto themselves a Utopia the like of which the universe has seldom witnessed. Thus tis unacceptable for us to sully their crystal aquas with the riffraff from the other side. Be selective with thine advocation. Those who appear to require detailed directions as to the whereabouts of the Start Menu should be immediately disregarded. Little can be done for those who cannot see beyond the limitations of the task basked user interface. Only those that climb in unaided over our built-in firewall warrant a place on our plage de paradise.

And may the Jobs be with you

iPaul