There are good reasons for a lot of the restrictions he's talking about (and "virtual machine software" is not an adequate solution), but he highlights one restriction that has no place in a modern enterprise: email storage restrictions.
Email storage quotas are such an outrageous false economy that it's surprising that the people suggesting them aren't drummed out of the company. It is not hard to figure out how much 10-15 minutes of someone's time is worth; divide their fully loaded cost by the number of work hours in a year.
Compare:
1 minute of admin/office work @ 0.35
1 gigabyte of disk storage @ 0.07
Companies should be looking for ways to shove more gigabytes down people's throats, in the hope of getting a 30-50% chance of an extra minute's productivity.
[note: the math doesn't even work out at NetApp's $/g]
Many companies offer intentionally small email quotas to encourage expiration of messages beyond the legal data retention requirements. Should you get sued, you don't want to reveal any more than absolutely required by law during discovery.
This is handled by date-based expiration, so policies I have seen say that "the company policy is 60 days after which email is automatically deleted". There is no reason to have a quota for this purpose.
If the company thinks old email puts it in legal jeopardy, why should the mail be kept at all? I hear this excuse all the time, and "excuse" is exactly what I think it is.
There's another reason for email quotas. The versions of Outlook in use by many companies use a PST (Outlook's data file format) format that totally breaks when it reaches 2 GB. Fixing it requires either truncating part of the file and forcing it to rebuild the internal structures or restoring from a backup.
I completely sympathize with this viewpoint. That said, I've been on the IT support side:
1. You have to support whatever is in use. Even users who say "you won't have to support me" end up needing support. This may be due to user error, or a host of other things like incompatibility with current network/firewall/computer policies. Best case, the new app doesn't work and the user ignores it, knowing it's unsupported. Common case, the use asks for help anyway "just this once". Possible case: the new app actually breaks something for other users (a quick example being generating abnormally high net traffic).
2. One big problem is the insistence on keeping all data in-house. This may or may not be more secure depending on a host of things. However, Joe Random user is not necessarily prepared to make the determination of "Is this company a reliable repository for my company's proprietary data?" (Yes, much of "proprietary" data isn't all that, but that's beside the point of this argument).
There are a number of companies that I would trust with sensitive information. Regardless, plenty of companies do banking via online portals, but reject putting other data in the hands of third parties.
3. Backups. This is a biggie. If I can't touch it, I can't be sure it's backed up.
But, man--I'm a corporate wage slave and these apps I have to use are like pulling teeth. I get that there are obstacles (see above)--but nobody seems to be doing anything to convince people/mitigate risk. And until the guy doing the purchasing can tell good design from bad, this isn't going to change.
The reason is that the entity (IT department) making the technology decision is optimizing for themselves. They will choose the product that minimizes work for them. They are not accountable nor do they care about the user's productivity.
Contrast this with an individual or startup, where the technology decision maker has a vested interested in its usability.
IT departments are trying to maximize aggregate efficiency across the organization (which includes themselves). While it might marginally increase productivity of office staff if they're allowed to choose whatever mail client they're used to, it is going to require a considerable investment in IT staff to support it.
Please keep in mind that an office full of programmers is absolutely not the norm. While programmers and engineers might hate the evil IT department, most offices wouldn't be able to exist without their mean policies.
most offices wouldn't be able to exist without their mean policies.
Man, talk about inflammatory. Offices have existed for centuries without mean IT policies. I'm not even sure how to read your comment in such a way that it models ANY reality.
I think it's a mistake to specify 'IT department'.
Just as IT departments often don't have usability as a primary concern, the corporation itself often doesn't have technical fitness as a concern. To them, almost all technology decisions are dominated by the concern of corporate risk.
They want industry-standards, because it keeps cog replacement costs down. They want you to use their gear to minimize data security concerns. They want to shut down youtube, lest they appear to be not enforcing the harassment policy. They force IE6 because they're trying to meet return projections on bad investments from years ago. They provide a standardized environment so they can shuffle cogs more easily. Etc.
Even if the average user always chose the best available tool, even if they were security-conscious, studiously making regular backups and vigilantly updating their software - corporate concerns would still regularly trump their desire to have X.
Just pick an IT guy and ask him how much say he has in what software he has to use all day.
You forgot the third goal: the IT department wants to create a lot of work for itself that appears to be useful but doesn't directly increase productivity.
If IT departments wanted to minimize work for themselves they would all outsource and use Google Apps. But then there would be vastly less work to do and as a result a smaller budget (since the money goes to Google instead).
Corporate IT today is a microcosm for the way established corporations become sclerotic and fail to compete with new businesses.
When you're building from scratch on a green field, it's easy to buy the most modern and efficient technology. Most corporations didn't build out their IT infrastructure until the early 2000s. Once you've invested in these systems, you're stuck balancing the cost of maintenance against the short-term cost of starting again, and the latter rarely wins.
This is why Windows XP and IE6 aren't going away anytime soon, but it's also the reason the last few steel mills in the US are going out of business with century-old equipment while cutting edge plants are built in south-east Asia.
One thing I often run into is that the financial people at our University regularly send Excel spreadsheets to one another. There will regularly be email chains of 3+ people that result in the same document being sent back and forth by email 10 times in one day. And then I get complains that people are hitting their email quota of 500MB.
The problem seems to be that you can convince individuals to switch to using a shared folder on a server (got my department using one) but they can't seem to make everyone switch. They've got email, it works well enough, and now it's hard to switch.
Technology can fix this! It was never the users problem to begin with...
MAX_ATTACH = 50 * 1024 #50 kb
def minimize_size(msg):
if size(msg['attachment']) > MAX_ATTACH:
upload attachment to s3 like storage
insert link into email body
return msg
It obviously needs fleshed out (I bet it ends up around 20 lines of python after everything is said and done). But seriously, the fact that your users would ever even need to think about this is just wrong.
If it's exactly the same file, mailservers (at least some) are clever enough not to store it for each instance. New features in ZFS are also, apparently, clever enough not to store two files in two places. But yes, shared volumes or Google Docs is probably the way to go.
Perhaps set the quota params so no [internal] message can be more than ~10k or so? If everything is forced to be in shared folders, users stop being defaulting to email.
I cannot stress how much I agree with this article. At work our corporate apps are all IE6 only, and even new machines purchased just last month were imaged with Windows XP. Outlook E-mail boxes were limited at 90mb. Speaking with the IT team proved fruitless - they were not interested with catching up with the times, or doing anything that required work on their part. As far as they were concerned, every XP machine came with IE6, and that's all they needed to ever support. Aggravating.
Corporate failures to make good technology choices should not be conflated, as they are here, with being unable to look at YouTube at work.
The fact of the matter is that 90% of office workers shouldn't be using their office computers for general-purpose computing. They shouldn't be stifled from taking initiative by the IT department, but neither can they be allowed to jeopardize the electronic security of the entire organization.
In short, the real problem is that many IT departments are complete failures, for a whole host of reasons. Duke University is the only large organization that I have ever encountered with a really well-run IT group. Two years after I graduated, management decided to outsource @duke.edu email to Sun. There was a nearly week-long unscheduled outage, and many people lost their email archives (some going back 10 years or more). (In fact, they seem to be having another serious problem right now): http://it.pratt.duke.edu/aggregator/sources/1?from=60)
Yes, I am so tired of seeing this. I've been listening to this user whine since the days of the Apple II vs the mainframe. Yes, tech advances faster than organizations can absorb. Eventually, the organization will catch up. Until people stop worrying about unmanageable IT costs, prevalent breaches, and dumb users this will never change.
I am part of an IT department and I would gladly move our 350 users to Google Apps, Gmail, etc. Google could do more to make their products seem less "mavericky" for the office environment. For one thing, it would be great if there were Google certified partners, the way MS/Orable/SAP has thousands of partners.
Wow, I rarely take Farhad Manjoo's side, but this IT guy sounds like he simply wants to avoid doing his job.
He even ruins his excellent point – that sometimes the real enemy is the management who set ridiculous policies for IT to execute – by calling it insane for anyone to go over IT's head and complain to, say, the secretary of state in a meeting held precisely for such questions.
I work at a University. Every time I have to work with one of our admins to get something through the Oracle purchasing app I feel like I just jumped back 10 years to Windows 98 and Web 1.0. It drives me insane how many things have to be clicked on to get a simple thing done and how it really looks like a very thin layer over their database.
So who is going to be the innovator who can supply the enterprise apps for 10k+ employee companies that actually work?
Email storage quotas are such an outrageous false economy that it's surprising that the people suggesting them aren't drummed out of the company. It is not hard to figure out how much 10-15 minutes of someone's time is worth; divide their fully loaded cost by the number of work hours in a year.
Compare:
Companies should be looking for ways to shove more gigabytes down people's throats, in the hope of getting a 30-50% chance of an extra minute's productivity.[note: the math doesn't even work out at NetApp's $/g]