Jump to content
House Price Crash Forum

Archived

This topic is now archived and is closed to further replies.

QP1

Why Do It/im Managers Have So Little Clue?!?

Recommended Posts

I've just come out of a meeting at work and feel the need for a rant...

I have developed a perfectly good database system for creating technical documents such that they adhere to certain standards and use standard terminology in key places; also the system allows for fast and sophisticated searching such as looking for text within sections of the document (e.g. when is chemical X specifically referred to as a risk, not just in the general body of the text). The system has a server backend and a windows application for the front end. Another site is interested in adopting the system but their IT/IM guys have a different vision and would like changes...

1. They want a web front end. When challenged why: "Because it will be better for the users". When does a web font end (browser based application) ever give a better experience than a native application? If you have a choice between OWA and Outlook, which do you use? If you have a locally installed word processor or spread-sheet, would you ever choose to use one of the web based alternatives?

There are sometimes good reasons to deploy applications in a browser but "a better user experience than a native application" is never one them.

2. They want the layout of the front end to look like the final report that comes out of the system, "because it will be better for the authors (users)". Why would it be better? Surely a system where you can quickly navigate between documents and between sections of your document and view the content by making optimum use of your screen is better than something that tries to mirror the final output (A3) and thereby makes you scroll and scroll around. Also, what happens if/when you modify the output layout of your data (documents) and they no longer match your input screens? You have to redevelop all your input screens to match your new report layout. What if, in the future, you want to be able to view your content on a 10inch tablet screen but your authors have tailored the text with spacing and line breaks that will only display correctly on web version and/or the normal A3 printed version?

3. Allow text formatting in free text fields. Invariably one person will use red to mean "safety critical", another person will use red to mean "obsolete text" and another will use red to mean "newly inserted text". Even if you could get everybody using red to mean the same thing, what are the odds that they will all choose the same shade of red if you give them complete freedom over formatting? And what about when the document is printed on a black and white printer and all the information highlighted as safety critical is indistinguishable from all the other text in the document? Or you plan to load your data into a system that does not support text formatting (e.g. pretty much every ERP system out there). Keep in mind too that a key deliverable is the ability to search the content (i.e. the meaning) of the documents and it is very difficult to search on "XXX" only if it is red, especially if we are talking about searching via a corporate document management/cataloguing system such as SharePoint.

In a technical document the author should be concentrating only on the content and not the formatting or layout - What-You-See-Is-What-You-Mean instead of What-You-See-Is-What-You-Get. Just make sure the meaning is clear from the text alone and you can then print it in whatever format you like or export it to any system you like without fear of losing any meaning. Simple.

How is it that a lowly developer like me can immediately see all these issues but a bunch of senior IM managers in a big company can hold multiple workshops (where they have plenty of time to think things through) and come up with such flawed ideas?

I didn't point any of this out as I think I'd rather stay schtum and get some work out of the situation (I'm a contractor after all) than make a fuss and have the big-wigs abandon the project because the dev is a stroppy-know-it-all... What would you do?

Share this post


Link to post
Share on other sites

tl;dr

But the answer to your question is because the people they report to generally have no clue whatsoever, aren't interested in trying to understand, and it is therefore easy to get away with being incompetent.

Share this post


Link to post
Share on other sites

I always wondered why they never asked the people that will eventually use the systems for their thoughts and how best things could be improved.....they ask their managers, then get it wrong because the managers think they know best when they don't always do, they don't use it......the developers don't care, if it fails they get another contract to amend what could have been got right first time.......that's not good for business. ;)

Share this post


Link to post
Share on other sites

Because people get to the top by being "almost good enough" and not rocking the boat.

I've been pulling my hair out at work over a similar issue. We've just bought a very expensive new imaging application; very sophisticated, 3D image processing, lots of clever tricks to speed up loading: 256-bit encrypted pre-caching, sophisticated hyper-dimensional/hyperspectral wavelet compression, etc. Fearing a uncoming staffing crunch, a key feature we looked for was the ability to support remote working, and the software does this admirably with intelligent pre-fetching so while you're working on one file, the next will be transferring, worklists, auto-syncing of files and worklists between users, etc.

However, IT managers have decided to enforce certain restrictions: as the data is confidential, no data can be cached on hard drives (even if encrypted). So, even though the workstations are on their own private gigabit subnet with the servers, the software loads the compressed data and decompresses it to the local hard-drive (except the local HD has been redirected to the user profile NAS, which is on a 100 mbit network). Unsurprisingly, it runs like a slug.

So, if encrypted data on an on-site hard drive was a problem, what came of remote working? The only option that would even be considered was to run the application on a terminal server, and connect to the terminal server using Citrix Metaframe. Now consider, that these are high-res, multi-dimensional datasets, where you need interactive 3D rendering and manipulation; that terminal servers don't provide 3D acceleration, and that the off-site upload bandwidth is 400 kbit - well, to compare it's performance to a slug would be doing the slug a disservice.

Not to worry, IT are now recruiting developers: Must have experience with ASP.NET, SQL including database administration, Post-relational databases, web design, JBoss/Jquery/JSP. £18k. I guess if you pay peanuts, you'll get monkeys.

Share this post


Link to post
Share on other sites

I see situations like this frequently, but don't have an answer. Once a project has acquired a critical mass of managment supporters, it will go ahead however stupid it is. As a contractor, you will probably have even less political influence than permie developers.

Have you discussed the system design with end-users? Their managers may only have a loose grasp of what is needed.

Having said that, I tend to be quite suspicious when anyone recommends a fat client Windows application over a thin client web front end. The support costs and cost of rolling out bugfixes and upgrades are usually much lower for thin clients. The recommender is usually a one of a herd of dinosaurs in my office who have never bothered to learn web development. Do I sound bitter?

In these situations, I try to influence the direction of the project where I can, but even misguided projects will pay my wages for a while, and hopefully I can learn new stuff.

Share this post


Link to post
Share on other sites
I didn't point any of this out as I think I'd rather stay schtum and get some work out of the situation (I'm a contractor after all) than make a fuss and have the big-wigs abandon the project because the dev is a stroppy-know-it-all... What would you do?

I'd do exactly this. I'm afraid after years of experience that it's pretty much guaranteed that these people would not listen so you may as well take the cash. The bigwigs can only be seen to be bigwigs if they make others do things only the bigwigs have thought of. They must demonstrate their power even if it means making things worse.

Share this post


Link to post
Share on other sites

They justify their own job by telling you to make changes to the work you have done. If it was perfect in every way functionally they would have told you to change the colour or the title. If the requirements they are now specifying were so important they should be part of the spec, if not you need the real users to give their feedback. I can guarantee the people who do the work day to day are the only ones who know the real impact of the changes.

It's why I always try and work by prototyping to prevent major rework (used to be called RAD until someone renamed it Agile after discovering big post it notes).

Seeing the amount of idiots, @rse lickers, back stabbers who manage to climb the chain in large organizations never ceases to amaze.

Anyone working in IT who says 'I'm not technical' should be sacked. How can you work in IT with no aptitude for it whatsoever. Like a blind airline pilot.

I feel better for that rant, thankyou.

Share this post


Link to post
Share on other sites

I quite often 'work in IT' - but am far from technical :lol:

Share this post


Link to post
Share on other sites

Well Mr Smartypants,

What colour should the wheels be?

(apologies to Douglas Adams).

Most IT is about spending vast amounts of money making things as complex as possible so that someone can cream off the support contract. Some large well known database companies make their money by releasing patches that cannot work and then hiring out consultants who charge to essentially rewrite the patch scripts so that they do work. It isn't just gubberments that get screwed over IT projects, most companies do as well.

Share this post


Link to post
Share on other sites

I once read that building contractors in California would deliberately put in a couple of features that were obviously against building regs.

The state official would come along, get all annoyed and officious, and shout at the builders to fix this crap. Which they did, because the mistakes, although they looked big, would actually be quite easy to fix.

The public sector drone was happy, because he'd met his quota of defects detected, and had his chance to look important. And the builders were happy, because they could f***ing get on with their job.

This anecdote might be of interest to anyone who has to present a new IT product to a conference room full of middle managers.

Share this post


Link to post
Share on other sites

I once read that building contractors in California would deliberately put in a couple of features that were obviously against building regs.

The state official would come along, get all annoyed and officious, and shout at the builders to fix this crap. Which they did, because the mistakes, although they looked big, would actually be quite easy to fix.

The public sector drone was happy, because he'd met his quota of defects detected, and had his chance to look important. And the builders were happy, because they could f***ing get on with their job.

This anecdote might be of interest to anyone who has to present a new IT product to a conference room full of middle managers.

Problem is the IT managers not only won't spot the flaws as flaws, they will identify them as the most essential features of the new project. Then they'll insist those features are included in all future projects and inserted into existing applications too.

Share this post


Link to post
Share on other sites

One reason is almost certainly because it's easier for their desktop estate people to install and there's a lot less comeback on them if Web based (buttcovering). IT Managers are generally in charge of all the monkeys that install all the software requests onto machines in a company either manually or via some automated system like SCCM. They have very little to do with nor do they much understand functional use of software. Amongst their main goals is, is it easy to install (i.e. in this case it would be easier for them if it was Web based). If it's Web based, they don't have to worry about whether it will work on admin restricted machines or getting their monkeys to install it manually at remote sites or reinstall it when they reimage a computer, nor do they have to consider creating a script to install it via GPO etc. (that is nowhere near as complicated as they want it to sound, it's still basic admin work). It's almost certainly nothing to do with end-user experience (you are of course correct, Outlook is much better than OWA).

2. Print Preview button? I don't get this at all.

Share this post


Link to post
Share on other sites

Because IT requires very little training in general.

You can do this job and be very successful with the following:

No maths skills

No language skills

No management skills

No personal skills

Just be good at troubleshooting, have thick skin, and have the ability to avoid any responsibility, blaming it on some problem out of their hands.

Most IT folk haven't a scooby about hardware or specific troubleshooting algorithms as taught by renowoned universities, back of matchbook poly's and internet schools.

Computer science is wholly different ball of wax of course, and I don't classify them as IT.

Share this post


Link to post
Share on other sites

Because people get to the top by being "almost good enough" and not rocking the boat.

I've been pulling my hair out at work over a similar issue. We've just bought a very expensive new imaging application; very sophisticated, 3D image processing, lots of clever tricks to speed up loading: 256-bit encrypted pre-caching, sophisticated hyper-dimensional/hyperspectral wavelet compression, etc. Fearing a uncoming staffing crunch, a key feature we looked for was the ability to support remote working, and the software does this admirably with intelligent pre-fetching so while you're working on one file, the next will be transferring, worklists, auto-syncing of files and worklists between users, etc.

However, IT managers have decided to enforce certain restrictions: as the data is confidential, no data can be cached on hard drives (even if encrypted). So, even though the workstations are on their own private gigabit subnet with the servers, the software loads the compressed data and decompresses it to the local hard-drive (except the local HD has been redirected to the user profile NAS, which is on a 100 mbit network). Unsurprisingly, it runs like a slug.

So, if encrypted data on an on-site hard drive was a problem, what came of remote working? The only option that would even be considered was to run the application on a terminal server, and connect to the terminal server using Citrix Metaframe. Now consider, that these are high-res, multi-dimensional datasets, where you need interactive 3D rendering and manipulation; that terminal servers don't provide 3D acceleration, and that the off-site upload bandwidth is 400 kbit - well, to compare it's performance to a slug would be doing the slug a disservice.

Not to worry, IT are now recruiting developers: Must have experience with ASP.NET, SQL including database administration, Post-relational databases, web design, JBoss/Jquery/JSP. £18k. I guess if you pay peanuts, you'll get monkeys.

You think you've got problems, I've run into a brick wall trying to get more hard drive space. Apparently it's OK for developers to spend hours or even days trying to find enough hard drive space to work with, but a couple of terabytes of hard disk would be far too expensive.

Share this post


Link to post
Share on other sites

Clients eh?...who needs em.

One recent story, way way below the OPs scale, was a small scale RD Terminal server installation.

How many users....3

What do they want to do with it....Get emails and run the database

Anything else....no

OK.....system goes live and being MS, it has 120 days grace period to see if you like it or not and to see if its suitable.

I call client regularly and ask how many licences they have decided on.....how long have we got left on the Grace period....60 days...30 days

5 days.....( the RDC tells the user how many days they have left.)

I never get an answer on how many users......although 5 regularly use it.

BOOM....the countdown is at zero....

BOOM, users a unable to connect.

I respond immediatly and of course have to ask how many licences they need....WOT licences?..you never told us...( lets forget the written quotes( plural) the summary report and reminders)

Its all my fault..

course the remote site cant get emails....what am I doing about it...well apart from having spoken to users and ensured they got them the "old" way, I wasnt aware they werent even doing that.

The database failed due to mods on the server by the new firm of offline webdesigners, but even though that database was the whole reason for having RDC in the first place, that didnt even come up.

Customers?....who needs em

Doesnt help that due to my accident, I didnt invoice the client fo 6 weeks...he prolly thought the IT was very cheap for a couple months...Hes in for a shock at this months Board meeting.

Oh, and the new all singing all dancing web based database designers experience was with small workgroups using a simple domestic router service.....our system has firewalls and stuff, and their installations wouldnt work.

My fault again......

Now, apparently, Im being paid a monthly fee to handhold them.....yeah right....as soon as i get an order based on my quote and its invoiced and paid.....no chance.

One has to wonder if sometimes these situations are due to wilful neglect by the managers.

Share this post


Link to post
Share on other sites

I'm not saying that some managerial decisions aren't insane, but...

as someone who trained as a developer and has since managed teams of developers, I can say that it's often frustrating to see developers hare off with some good idea without thinking about the consequences of delivery.

It's a bit like law. Laws are easy to make ("Let's just make tax evasion illegal! Let's ban dangerous dogs!") but pretty hard to implement well without screwing lots of other things up. This is why these blocking processes and committees gain traction.

I get very frustrated when I can't do something that's obviously good for the client, but often they don't want "quick, cheap and effective", they want something they think of as stable, predictable and reliable. Whether they make more money or not is often not their main consideration.

Share this post


Link to post
Share on other sites

You think you've got problems, I've run into a brick wall trying to get more hard drive space. Apparently it's OK for developers to spend hours or even days trying to find enough hard drive space to work with, but a couple of terabytes of hard disk would be far too expensive.

I don't want to worry you, but the last time I experienced that sort of apparently illogical behaviour was shortly before the company went pop. In our case it was a RAM shortage. Some of the developers had even taken to buying and fitting their own memory modules in order to be able to do their job properly. Of course, unknown to us, the decision had already been taken to shut down the company and no more money was to be spent on anything other than unavoidable expenses (such as salaries) in the remaining time.

Share this post


Link to post
Share on other sites

I don't want to worry you, but the last time I experienced that sort of apparently illogical behaviour was shortly before the company went pop. In our case it was a RAM shortage. Some of the developers had even taken to buying and fitting their own memory modules in order to be able to do their job properly. Of course, unknown to us, the decision had already been taken to shut down the company and no more money was to be spent on anything other than unavoidable expenses (such as salaries) in the remaining time.

Thanks for that..

Actually, we are hiring a couple of people at the moment so I'm reasonably sure about my job for at least the next couple of years (is that the sound of fate being tempted..?) They've been a bit illogical about hardware for years.

Share this post


Link to post
Share on other sites

Clients eh?...who needs em.

One recent story, way way below the OPs scale, was a small scale RD Terminal server installation.

How many users....3

What do they want to do with it....Get emails and run the database

Anything else....no

OK.....system goes live and being MS, it has 120 days grace period to see if you like it or not and to see if its suitable.

I call client regularly and ask how many licences they have decided on.....how long have we got left on the Grace period....60 days...30 days

5 days.....( the RDC tells the user how many days they have left.)

I never get an answer on how many users......although 5 regularly use it.

BOOM....the countdown is at zero....

BOOM, users a unable to connect.

I respond immediatly and of course have to ask how many licences they need....WOT licences?..you never told us...( lets forget the written quotes( plural) the summary report and reminders)

Its all my fault..

course the remote site cant get emails....what am I doing about it...well apart from having spoken to users and ensured they got them the "old" way, I wasnt aware they werent even doing that.

Should have used Linux and open source software, no need for licenses means the customer won't experience license issues and can spend more money on your consultancy rather than paying the Microsoft tax.

Win win for both of you, happy customer and happy Bloo Loo.

Share this post


Link to post
Share on other sites

A friend of mine is a manager in an IT company. He started his career as a coder and is a very clever guy. He describes his fellow managers as 'idiots who know nothing'.

Share this post


Link to post
Share on other sites

Thanks for that..

Actually, we are hiring a couple of people at the moment so I'm reasonably sure about my job for at least the next couple of years (is that the sound of fate being tempted..?) They've been a bit illogical about hardware for years.

Companies hate buying "stuff" as it's seen as a "cost", whereas paying developers is something that can easily be removed to make cash flow look better.

It makes no sense if you're looking at overall resource use per se. Unless they want to discipline people into using resources properly. There's some logic to this, as hard drive additions often increase other costs, and are mostly full of crap anyway.

Share this post


Link to post
Share on other sites

I have developed a perfectly good database system for creating technical documents such that they adhere to certain standards and use standard terminology in key places

Wotcha doing re-inventing the wheel? Unless your company is doing something oddball aren't there zillions of suitable DMS / doc standard systems already available for not a lot of money?

Share this post


Link to post
Share on other sites

1. They want a web front end. When challenged why: "Because it will be better for the users". When does a web font end (browser based application) ever give a better experience than a native application? If you have a choice between OWA and Outlook, which do you use? If you have a locally installed word processor or spread-sheet, would you ever choose to use one of the web based alternatives?

There are sometimes good reasons to deploy applications in a browser but "a better user experience than a native application" is never one them.

There are still some apps which have bandwidth and latency needs which can't be satisfied easily with browser based apps. Like for example a high volume realtime EPOS terminal. And I'd hate to see a web-based version of Visual Studio, even the PC based one runs slowly enough sometimes as it is. Photoshop is another one which doesn't really work, we just don't have the sort of bandwidth needed to be able to "remote" everything.

But although I started out building Windows apps I haven't created one in a decade. The ability to simply call the app up in the browser and use it immediately without having to install anything ("Can I use your machine for a minute, I don't have X installed on mine") pretty well trounces almost everything else as does the ease with which you can then access the application outside the office from anywhere if you set it up like this.

The downside is that browser based apps and printing don't go harmoniously together.

2. They want the layout of the front end to look like the final report that comes out of the system, "because it will be better for the authors (users)". Why would it be better? Surely a system where you can quickly navigate between documents and between sections of your document and view the content by making optimum use of your screen is better than something that tries to mirror the final output (A3) and thereby makes you scroll and scroll around. Also, what happens if/when you modify the output layout of your data (documents) and they no longer match your input screens? You have to redevelop all your input screens to match your new report layout. What if, in the future, you want to be able to view your content on a 10inch tablet screen but your authors have tailored the text with spacing and line breaks that will only display correctly on web version and/or the normal A3 printed version?

Users often seem to find it hard to differentiate the content from the layout. A previous boss used to be like this - you can't write a single word of text to go on a page unless you have the template 100% done and the text is the only and last thing to add. Except as you say, that approach has been blown away by the variety of screens and devices now making it essential to separate content and layout.

I don't think there's a "right answer", how about a Print Preview button? (Mind you I've been asked for one of those even when the screen exactly resembles what the printed copy will look like anyway, or when the print is something like an invoice which you don't really need to "preview", do you, it's not a piece of craft work, it will look like the table on the screen with the logo at the top and the totals at the bottom the same as it always does...)

3. Allow text formatting in free text fields. Invariably one person will use red to mean "safety critical", another person will use red to mean "obsolete text" and another will use red to mean "newly inserted text". Even if you could get everybody using red to mean the same thing, what are the odds that they will all choose the same shade of red if you give them complete freedom over formatting? And what about when the document is printed on a black and white printer and all the information highlighted as safety critical is indistinguishable from all the other text in the document? Or you plan to load your data into a system that does not support text formatting (e.g. pretty much every ERP system out there). Keep in mind too that a key deliverable is the ability to search the content (i.e. the meaning) of the documents and it is very difficult to search on "XXX" only if it is red, especially if we are talking about searching via a corporate document management/cataloguing system such as SharePoint.

Thinking in "web mode"..

I wonder if what might work is a rich text editor with a bit of customisation to regonise custom markup and a couple of buttons to invoke it so you end up with:

<newlyinserted date="ddmmyy" time="hhmm">new text</newlyinserted>

in the source view, but red text in the Html view? So the text concerned preserves meaning and it's always the same shade of red since the system colours it in.

Then you could do a search for "all the articles in the last month with newly inserted text" whether on the document itself or with a "mart" field called LastNewInsertion at the document entity level which gets populated on save..

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

  • 239 Brexit, House prices and Summer 2020

    1. 1. Including the effects Brexit, where do you think average UK house prices will be relative to now in June 2020?


      • down 5% +
      • down 2.5%
      • Even
      • up 2.5%
      • up 5%



×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.