As an IT manager for a smallish business I am so disappointed that the 16" m1 max doesn't support external ram or SSDs. I have tight budgets and need to make machines last 3-5 years. On the PC side I standardize on laptops with slightly over powered CPUs, but typically right size the ram and SSD for what I expect our users will need around year 3. If a user changes jobs and needs more resources, or our tool chain changes drastically for a team, I just order new ram and SSDs and upgrade. If a user is still fairly happy with their machine nearing the 3 year mark I can usually give them a boost with some more ram or storage and get another year or two out of a system for a fraction of our yearly amortized average for a new machine. Without an enterprise class machine that is meant for midcycle upgrades it means I have two choices, right size for where I expect to be in 3 years and risk a huge upgrade cost if something changes, or pay through the nose to max out every option and triple my yearly per user hardware budget. Looks like next years hardware refresh might be all PCs.

Call me cynical, but if I was a large organization with reasons to hide my inventory information from competitors, but also reason to release info to the public about the services I am providing. And if I could get the cost of my proprietary, and identifiable, add on board down to US$10 at scale. Then every time I brought something in for repair I would swap out the that component and just have a table in my database that keeps track of ID changes. From the outside world it would look like I have deployed way more devices then I actually have. And it might trick more naive competitors into thinking they need to replace their hardware more often. Driving up their costs and increasing their chance of failure giving me more market share opportunity in the future, or maybe give me more leverage in negotiations with regional authorities.

...that you don't want to be flooded with offers from random internet users. But, your best bet is to provide more information, such as the general area (local major city/state/country) and ask a group of experienced IT people for recommendations for a local IT guy that would be interested in a support consultant gig. Kind of what you are doing now, but with more detail. For a business of this size they are better off finding a local IT guy with a regular full time gig that wants to make a few extra bucks supporting small businesses on the side. I do exactly this in my area, and all of my work comes from recommendations from other technical people who either don't have the time, or don't want the relative frustration of acting as a consultant.

This has been an open secret for years. They want to reduce their higher paid headcount by 100, so they lay off 1000, but they give the employees time to apply for other internal jobs. But make sure that all of the open positions the laid off staff are eligible for pay less then the salaries of the employees you want to get rid of, and higher 900 back in different rolls. This same scheme works to get rid of people getting close to retirement as well, at least it did back when IBM employees still had pensions.

All external storage has some volatile receive buffer space in ram, or even on chip cache. And while 1 and 2 in your list are likely unimportant, for the most part, the biggest problem is if you disconnect power before the device has a chance to optimize the data and write it out to the persistent storage. Depending on the speed of the storage controller, how big the on-board buffer is, and how much work it is doing to optimize the write, and how fast the actual write can be performed, it can take several hundreds of milliseconds or more to finalize the last write operation. Look at modern spinning drives and you will see some that have a few hundred megabytes of cache, and it can take several seconds for the cache to be written to disk even after the OS believes the transfer has completed. The only real solution is to add rechargeable power backup to the device so that you can guarantee that even if the drive is disconnected it can finish writing the data in memory to disk before it shuts down.

The truth is that if a device manufacturer is hiding a background process out of sight of the user and the OS then it should be built with suitable backup power, in the form of a small battery or ultra capacitor, to guarantee that the data in ram is flushed to persistent storage. I understand that this would be easier to do with flash drives then external spinning disks, but you could still probably make a spinning disk reliable with a small lithium ion cell, like a 14500, hidden in the case. Any "routine" or "common sense" procedure that can be eliminated with a simple hardware fix is not in the best interest of the end users. It is the same reason microwaves automatically shut off when you open the door instead of having a warning in the manual that you may suffer microwave burns if you fail to turn off the microwave before removing your food.

One can dream right?

Since you own one, and I can't find mention of it on Seiki's site. Does this monitor support split screen from two sources? would be nice if it could handle 2x 60hz inputs instead of being locked at 30hz.

Alright, regardless of your take on FTDI's actions. Isn't the real problem here trying to fix a broken market with a regulatory or software solution?

I mean why is the FTDI chip so routinely copied, or cloned? It all comes down to price and availability. We saw this with online music, and we are seeing a corollary here. In this case the end users aren't the market though, they are collateral damage in the dispute between FTDI and hardware manufacturers. FTDI has a product the market wants, but they are asking for a price that the market doesn't want to pay. So people are stepping in with drop in replacements for the parts that the market doesn't want to spend money on, or can't get access to. The best way for FTDI to fight clone makers is to lower their prices and raise their production until the market decides that taking the risk with a clone chip isn't worth it.

And this doesn't just apply to FTDI. This applies to anyone making a commodity part that is widely used by the electronics industry. Sure high quality part manufacturers will never be able to bring their cost down the exact same level as cheap knockoffs, but if they get closer they will recapture some of the market. In the case of the FTDI chip we are discussing right now, the part has been on the market long enough that they have probably made up their manufacturing and tooling costs at this point and could lower the price to meet market demand if they wanted to.

Maybe we just need to push for a "generic" chip industry similar to the US drug market, though the protected window for the original designer would have to be much shorter to factor in the shorter dev/test/to market lifecycle of electronic components. By this time would could have authorized FTDI usb to serial clones, and FTDI would be banking a fraction of a cent per unit while working on the next faster or more power efficient model.

I built a rig like that with 3x22" monitors in portrait mode. Build my own mounting system for them using MDF and plumbing parts. It does kick ass for games. But productivity goes way down for anything else because interfaces like web pages, IDEs, Email clients, etc. are all built for 16x9 not 9x16 so the bezels break visual consistency.

Alright, I have been trying to figure out who to gripe about this for weeks now. And low and behold this post hits good old Slashdot.

My wife signed us up for Progressive's "Snapshot" about 4 months ago, and I have been driving around with the under dash device since it arrived. We have about 2 more months to go. Here are my impressions so far.

1> The device only seems to care about breaking rates. It gives an audible beep whenever you decelerate at greater than 7mph/second, without the ability to log where you are it can't correlate speed limits or traffic patterns.
2> It does not seem to take into consideration terrain or slope. I can safely decelerate at greater than the require speed when going up hill, and frequently do.
3> it does not take into account state recommended yellow light timing. Here in the state of Texas our Department of Transportation recommends yellow lights last 1 second for every 10 miles per hour of the speed limit. This means that they expects us to be able to safely decelerate at 10mph/second. But they also suggest yellows last no more than 6 seconds, meaning on roads with a speed limit of 70mph or greater the expected rate of deceleration is even greater. So either I need to be precognitive and start breaking before the light turns yellow, or I decelerate at 7mph/second and end up somewhere past the light that I am stopping for.

This leads me to one conclusion. This device is not intended to benifit consumers. It is a thinly veiled Pavlovian training device to reduce accidents, benefiting Progressive. But since the "safety" standards are so far off of regulatory recommendations it is nearly impossible for anyone to actually meet the standard and get the promised discount.

I have considered contacting the state insurance board, or a class action lawyer, but I don't know that either would help.

Our VPN is set up as a realy, so our clients get handed DHCP leases with the DNS server and domain information included from an interal server. But occasionally we see a client who's machine will not honor the DNS servers given out by DHCP. So we us an internal subdomain (i.ourdomain.com) with all of the internal hosts listed there (like private.i.ourdomain.com). The NS record for the i.ourdomain.com subdomain points to an internal IP address so a user can only resolve internal hosts when connected to the VPN. So when a user tries to connect to an internal server across the vpn, DNS client follows the path (root servers->DNS server for ourdomain.com -> DNS server for i.ourdomain.com) to our internal dns server and then receives the correct internal IP address for the server. If they are not connected to to the VPN then they can't get resolution off of our internal servers and the lookup fails. This prevents us from having to publish internal DNS globally for clients that don't respect our DHCP load.

So to reiterate and simplify many of the previous posts before I ask my question; Copyright protects development of good or services (preventing loss of investment) and Trademark protects from identity theft. Now that we understand that. So if we equate brand identity to personal identity (true not perfectly equal but close enough for this discussion), then didn't we solve, for most use cases, this problem long ago by the use of sir names in human culture? So the obvious solution is that a trademarked name created by the products original developers could, if the developer opted in or the law was changed, be carried on by subsequent generations forked from the original as long as they included a unique extension. I.E. Firefox, could be forked by another team to bring you Frederick, son of Firefox, or something else suitably ridiculous. And the brand confusion argument would hold little water because we are preprogrammed to understand this nomenclature due to our social structure.

I have used both in the past with little discernible difference in performance for general computing tasks. But here is what my experience has taught me, neither is practical for _my_ needs . In my case I travel with my laptop a lot. When I first decided to go wireless I went with a cheaper proprietary RF mouse, even though my laptop supports bluetooth, because I am a cheap bastard. But after my second snapped dongle, one my fault as I had forgotten it was there when I threw my machine in my bag, and one because someone decided to squeeze by me in a coffee shop and snapped off the usb connector, I decided to go bluetooth. While my logitech bluetooth mouse did have a hefty price premium it has outlasted three laptops and doesn't require a proprietary dongle. My biggest complaint with both solutions was power consumption. In my experience power consumption was a function of implimentation. While my logitech unit is great sitting on a desk going into sleep mode almost before I let go of the mouse, it had no off switch. So if I didn't take the batteries out before throwing it into my bag it was dead in a matter of hours. While one of my RF mice did have an off switch, the sleep mode had a very long delay before kicking in and limited it's overall battery life to about a week. On a semi related tangent. when I worked for a large company management decided to standardize on on vendor/model for wireless keyboards and mice. But the RF model they chose had a limited "key" set so once an office reached critical mass we where inundated with call that started "Help! Someone is taking over my computer. I think we are being hacked." Only to find out that there wireless dongle was just associated with another keyboard and mouse set utilizing the same key. Because of the pairing requirements this should never happen with a bluetooth unit. I solved these problems with two approaches. For my bluetooth mouse I installed an off switch in line with the the positives battery lead. Now my batteries last me a couple of weeks even when traveling and I can still move the mouse to any machine that includes bluetooth. For the RF mouse, the one with the lost usb connector, I cracked the case to one of my old laptops and soldered the dongle to a set of unused usb pins on the mainboard. Now that laptop, currently used by my son, has a dedicated wireless mouse with no risk for dongle damage. In the end it all comes down to personal preference. What are you going to be using it for? How much are you interested in spending? And, what environmental factors would limit your options.