It's Cool for Me: Comments

Comment on "Building on the past. Designing for the Future."

Fri, 20 Nov 2009 17:44:20 -0500

Looks like a great plan, Chris.

Comment on "Spending Money and Not Communicating at the Same Time"

Fri, 21 Aug 2009 17:22:33 -0500

In the book Mavericks at Work by William C. Taylor and Polly G. Labarre, they tell the story of Corporate Lingo at Commerce Bank. The phrase I like is, “One to Say Yes, Two to Say No.” This means that every employee is empowered to say yes to a customer, but to say no, the employee must first check with his or her manager (two to say no). It seems to me that this is "lingo" that we should incorporate in ITS. It means leaving a culture of "No!" or "Yes by Committee" behind.

- Jim Leous

Comment on "Spending Money and Not Communicating at the Same Time"

Fri, 21 Aug 2009 16:29:36 -0500

Hmm, I checked a different route views server, and I can now see a v4 path from 3rox. I still don't see a v6 path from magpi.

- Derek Morr

Comment on "Spending Money and Not Communicating at the Same Time"

Fri, 21 Aug 2009 16:08:27 -0500

I don't know if this advertising is intentional. I can ask Justin about this next time I wander down the hall.

- Chris Kauffman

Comment on "Spending Money and Not Communicating at the Same Time"

Fri, 21 Aug 2009 15:59:25 -0500

It appears that we have some bugs to work out with the new routing system. From what I can see, we're advertising our IPv4 prefixes only to MAGPI and our IPv6 prefixes only to 3rox. Is this intentional? Or should I email the NOC?

- Derek Morr

Comment on "Spending Money and Not Communicating at the Same Time"

Fri, 21 Aug 2009 15:52:40 -0500

The President's Office of Planning and Institutional Assessment seems to say just about the same thing — albeit, less specifically about voice services — here in the May 2009 issue of Innovation Insights:

Remember that organizations are perfectly designed to get the results they get If you want to change the results, you need to change the system. As the saying goes, “If you always do what you’ve always done, you’ll always get what you’ve always gotten.”

http://www.psu.edu/president/cqi/innovation/creating_a_culture.pdf

- MARK H LINTON

Comment on "Dear Santa,"

Tue, 07 Oct 2008 11:28:32 -0500

Dear Chris, Mrs. Claus took your list from Santa. She said it was too long. She would like you to work with Santa to rank said items. Signed, Management

- Nikki Massaro Kauffman

Comment on "Blogging is hard so I will make it harder"

Fri, 05 Sep 2008 12:03:56 -0500

Notice I said more frequently than twice per month, but not every day. :-)

- Chris Kauffman

Comment on "Blogging is hard so I will make it harder"

Fri, 05 Sep 2008 11:51:47 -0500

Love this new idea—it takes a special kind of masochist to make frequent blogging harder. I'll be rootin' for you!

- rb smail

Comment on "Yellow Lines and Dead Armadillos"

Tue, 19 Aug 2008 16:56:18 -0500

I can relate to your concern on the dedicated team. I will trust you when you say it applies to the case at hand, but it is really a systemic issue. We all have full-time jobs. Otherwise, we would all be somewhere else. When there is something "new" to do, it is mostly volunteer time. The "team" you create only exists because you made it. That makes you dedicated. That is wonderful. An employer can say, "You are an exempt employee. You should do what it takes to get the job done." They would be right. However, it would be better if there was an institutional commitment to the team. We talk about project management strictly in a scholarly sense. It does not map onto any formal structure within our workplace.

- MARK H LINTON

Comment on "Yellow Lines and Dead Armadillos"

Fri, 15 Aug 2008 14:18:36 -0500

That's a good loop to include John. What I had in mind for managing expectations is that since this is new software, and software "always" works as advertised, we should be prepared for disruptions. I would distinguish disruptions from interruptions in this way. A service disruption would be changes to menus or web interfaces that we have gotten used to seeing a certain way. A service interruption which I consider unacceptable would be a complete loss of calling capability.

- Chris Kauffman

Comment on "Yellow Lines and Dead Armadillos"

Fri, 15 Aug 2008 13:49:54 -0500

You left out the side loop under "Manage Expectations" that goes back to re-Defined Goals. In this case, there used to be a goal of stability at every step in the process, and that is gone now. The management occurred in the seldom-used path of "Empowerment" where you told your boss NO - it can't happen that way. Fortunately he listens to reason. Also under Dedicated Team, you need to consider adding "kill off the opposition", or "find a cooperating vendor sales force", or whatever applied to what you did on the phone call earlier this week. :-/

- JOHN DOUGLAS BALOGH

Comment on "Why does Penn State use real IP Addresses?"

Tue, 05 Aug 2008 10:38:11 -0500

I don't think it's a policy, just an entrenched practice. Historically, public v4 addresses weren't scarce, so we used them wherever possible. There are two arguments to use private addresses: security and scarcity. I think the security argument is false. If you're using NAT, it's possible to tunnel through (some) NATs, and many attacks don't require the attacker to be able to open a connection to the victim. If you're not using NAT, you can still get to private addresses from inside the PSU network, and I think our network is more porous than many believe (how many people migrate laptops, smartphones, etc from home to work and back again on a daily basis?). The only way to prevent these sort of attacks are firewalls, anti-virus/spyware tools, etc. If you're using all of those, why use private IPs? Further, it's not possible to proxy all protocols. And some protocols do nasty things like embed IP addresses into various data structures (such as Kerberos, FTP and SAML, to name three off the top of my head). You can certainly mitigate these issues, but it's extra work. Using private addresses doesn't buy you much extra security, and it restricts what you can do. Frankly, I think it's more hassle than its worth in many cases. Certainly we need to keep it around, since so many people seem to want it and since PCI-DSS requires it. So that leaves us with the v4 address scarcity argument. I'll make two comments here. One, PSU has a sizable amount of unused IPv4 addresses (large portions of the two /16s that TNS does not control - CSE's and Hershey's). Two, IPv6 is the long-term solution to this problem, but it will be a very long time coming.

- Derek Morr

Comment on "Why does Penn State use real IP Addresses?"

Tue, 05 Aug 2008 09:25:22 -0500

Is it a policy? I think it's a convenience, at least. If you want any inbound traffic to your network that wasn't initiated from within, NAT just adds more hurdles. NAT isn't needed for protection; firewalls do just fine with public addressing. Housing could use it on the student networks, as they block all incoming anyway. That's about the only area where I see a blanket "Use NAT" to be reasonable. As far as running out of IPv4, well, NAT helps, but I would hate to use it as an excuse not to adopt v6 more quickly.

- Bill

Comment on "Why does Penn State use real IP Addresses?"

Tue, 05 Aug 2008 08:56:12 -0500

I think you're on to something here, Chris - but do you know the story about the apes in a cage with stairs leading to a banana? If not, I'd be happy to share it with you...

- Paul Carlisle Kletchka

Comment on "I, Robot"

Fri, 01 Aug 2008 20:29:31 -0500

Robot dreams effectively crushed by the promise of saving for Disneyworld. Nobody can program a robot the way Mom can program the kids. Sorry, Love.

- Nikki Massaro Kauffman

Comment on "I, Robot"

Fri, 01 Aug 2008 16:06:10 -0500

Lego robot is less salty, and works great for both li'l kids and big kids. My kid's uncle bought them one so I get to play too (when the kids let me :-).

- JOHN DOUGLAS BALOGH

Comment on "Voice Mail Dead? I wish."

Tue, 08 Jul 2008 14:38:16 -0500

I see a lot of "you" in your descriptions of UC control metrics; adding "supervisor" in there is probably required to actually make it happen. Otherwise "you" will never get funding for it and nothing moves forward. Even the WikiPedia definition misses that point. UC is a combination of supervisor control and user preferences. Some call it policy, others call it control, but it reads the same in the end: stuff you can not set (even if you think you know better). Don't get me wrong, I want to get an SMS when my wife/boss leaves me a voice/email, but I don't want to waste that money if you (or anyone else) calls. If the policy allows it, and the supervisor has granted that function to users, I'll gladly turn it off if it does not work the way I want it to work. Likewise I may subvert a call to another system entirely if I can control the notification through Jabber so that my MWI, Email, and car alarm all clear at the same time... (No, I don't have a car alarm, so clearing that bit will have no effect. :-)

- JOHN DOUGLAS BALOGH

Comment on "Voice Mail Dead? I wish."

Tue, 08 Jul 2008 12:39:20 -0500

Check out http://www.smithonvoip.com/voip-commentary/if-voicemail-is-dead-then-the-phone-call-is-dead/ "Force us to unitask" - GOOD. Multitasking is the Great Lie. The truth is that humans are terrible at multitasking. If we were to focus more on one task at a time, including listening to voicemail (as opposed to listening to VM on speaker while reading e-mail or the web) we'd probably make fewer mistakes and be less stressed-out. I'm personally indifferent about voicemail. But if you leave me one that is longer than one minute I'm likely to delete it after the one minute mark--regardless of whether I'm listening to it on my phone or as an e-mail attachment. Get to the point, people! (Similarly, VMs that say "Call me back" without giving me a reason to do so are likely to be ignored. :-) )

- Bill

Comment on "How can I justify this project?"

Tue, 17 Jun 2008 18:59:36 -0500

SMS to your cell phone works just about as well. I liked that the funding came from a Windows prize, but they only used Linux in the demo. :-o

- JOHN DOUGLAS BALOGH