Posts by gtarthur

The cybersecurity QA trifecta of fail that may burn down the world

Re: The road to hell...

A good start is right here in this forum. The up and down voting of participants is a good start. Beyond that, is the example set by the venerable techie site slashdot.org. In addition to the voting, a user can limit what they see based on the net value of the up and down voting. This generally keeps trash below the cutoff threshold. I do like the categorization as well. It would help with those crazies that are either off topic or psychotic. As others have pointed out, the problem is the algorithms are tweaked to generate "hits" for advertising revenue. The failure to filter is also probably a function of "after the fact" content analysis (assuming there is any). Categorization should take place at the time of submittal. This has a 2-fold benefit. First it's a clearly identifiable event in the workflow, and second it introduces a delay factor that can work to slow down the rage machine. I have personally given up on all social media and focused on this type of forums and feedback for new journalism from "authentic" publishers. Perhaps this will resolve itself with a generational "turnover". Let's hope we don't burn it all down before that.

Energy buffs give small modular reactors a gigantic reality check

Re: Colour me cynical

That was exactly what I was thinking too - evaluating 4 devices that are based on technology that's over 10 years old is ridiculous. At the time those projects were starting, the only extant technology was naval propulsion systems. Previous commentors have correctly spotted the weaknesses of adapting those technologies for civilian power production. It's far too soon in the life cycle of SMRs to do this kind of "hit job" based such a small number of devices. At this point there's no way to evaluate the possible scaling factors that might be evolved.

Also, I would not grant any credibility for results from any of the 3 countries mentioned.

Intel has driven a dagger through Microsoft's mobile strategy

Learn to distinguish war(s) from battle(s)

My 35 years of experience in computing tells me this is yet another battle that may be lost, but is probably just a strategic retreat. How do I know - follow the money. Who has it to spend, and who doesn't? The big players will continue to be big because they know how to stay big - with both talent and cash. If Intel can't build an internal mobile chip, then they can buy someone that does. Failure for these guys is an acceptable and manageable risk. Just like the startups, the big guys can have failures that they learn from and keep growing. Too soon to bury anyone in this war.

Why Tim Cook is wrong: A privacy advocate's view

It was not a personal phone

In the rush to marshal all the forces of good and evil, the main combatants have overlooked a glaring mistake by the terrorist's employer. The local government agency issued him a "company" cell phone without a mobile device management, MDM, agent installed that would allow the local government to take control of the phone. He didn't buy or lease this cell phone. It was provided to him, ostensibly, as part of the responsibilities of his job. Although in many agencies and companies, cell phones are merely perks, and the resulting lack of proper administration is clearly obvious in this case. The use of an MDM agent wouldn't have broken Apple's excellent encryption regime, nor violated the privacy of the phone's "owner" - because the owner was not the user, it was the local government. Once again lack of foresight and planning have created a social dilemma that could easily have been prevented in this case. As for privacy versus security, I remind everyone that each of us has the primary responsibility for our own privacy, and the first lesson is to keep it to yourself. Your phone is not an appendage.

The enterprise IT landscape: Five key ways it will change

Black box not white box

My humble opinion is that the secret here is that cloud is sold as a service, which removes all the second guessing and kibitzing from both within and outside of IS. Too often IS managers get drawn into compromises on internal systems to conform to the legacy kit already on site. Cloud makes it a business service decision. That allows the cloud vendor the greatest latitude to architect and operate a solution which meets both the internal business costing model and the business needs of the customer. Most large businesses are so weighed down with legacy and internal politics that only the most disciplined can ever manage to do this themselves. I would argue that only those companies that are themselves highly technical ever really manage to do this. The rest of the world is the reason that out sourcing goes through period cycles, and cloud is the name of the latest variety.