I replaced my Nextlink Bluespoon with a Plantronics M3000 headset. My experience thus far has been positive.
It is, unfortunately, larger than the Bluespoon and I find it less comfortable. I also appreciated the ability of the Bluespoon to completely block all ambient noise in the one ear so that I could easily hear the conversation--a feature I know some other people didn't like.
On the other hand, the M3000's microphone seems to offer better fidelity (although there is a slight humming that is noticible on my recorded sounds, but this shouldn't be noticed on actuall calls?). The M3000 is also usable in both ears and (my reason for picking it:) has a user replaceable battery. I have not been able to find another headset that has a user replaceable battery. It was also only about $65 from where I purchased it (retail, from a reputable store) and comes with a one year warranty. The battery is supposed to last approximately 12 months, according to its documentation ("for optimal performance").
I'm currently taking the ear loop off while storing it for travel. It fits better this way, and I often switch which ear I wear it on anyway. It is also very comfortable when worn upside down. Right side up, or down, it is much better than I expected. It is supposed to be put on with one hand, and to some extent this works (especially upside down), but sometimes I have to fiddle with my ear lobe to make it stay on when putting it on 'correctly'.
As I mention on the project page, I'm thinking of expanding the breadth of my thesis from a simple comparison of FreeBSD's implementations of Soft Updates and journaling file systems to a more generic examination of several file systems. I do not know if anything conclusive will manifest itself, but I would like to see empircally what types of file systems perform better at what, and by how much. This could compare the performance and reliability impact of Soft Updates over regular UFS and the performance and the performance differences between extent-based and B+ tree based systems with traditional bit-mapped file systems.
This of course is going to enter a lot of grey area, considering in many cases this will not be comparing 'apples to oranges'. Considering there are several systems based on similar technology (XFS, JFS, ReiserFS are all extent based, journaling systems based on B Trees [iirc, I'll need to check this again]) some trends should be possible to spot. There have, of course, been several small studies done comparing system to system, but I hope this will be a comprehensive analysis taking the underlying methodology into account and not just the systems themselves.
R.P.I. is playing Princeton this Friday at RPI. Season ticket holders could have ordered them yesterday but general admission tickets go on sale tomorrow at 0900. Unfortunately, my tickets are in someone else's name, who doesn't want to go, so I have to wait and see if there are any seats available. If not, I guess I might as well go home. It's irritating that I can't get my season seats, or really anything good.
I'm going home anyway this weekend, so if I can't get good seats I'll just blow it off. My sister is coming home for spring break, so I'll get to see her again. Next weekend I'm visiting Brown which is probably where I'll be for the next few years.
Either I broke something in my version of squirrelmail (very possible) or it doesn't like nexted multi-part messages (it give seach the same boundary marker). I need to do some further checking to see if this is the case. I spent a day working on the GPG plugin and didn't get it finished. I hope I'll be able to take that up again later this month. I need to see if I broke this though ....
--