Topic: Problems with an impatient boss and insane deadlines (Page 1 of 1) Pages that link to <a href="http://ozoneasylum.com/backlink?for=29177" title="Pages that link to Topic: Problems with an impatient boss and insane deadlines (Page 1 of 1)" rel="nofollow" >Topic: Problems with an impatient boss and insane deadlines <span class="small">(Page 1 of 1)</span>\

 
paritycheck
Bipolar (III) Inmate

From: you tell me
Insane since: Mar 2004

IP logged posted posted 05-02-2007 13:24 Edit Quote

Hi guys I'm really in a bit of a pickle here,

After lounging over a year for a job I finally landed one and have been working here for almost a year. The problem is that I'm the only Php guy or rather I'm the only IT guy save the network admin who has absolutely no idea of web development and my boss has these long list of projects for me to work on. Now my boss is a bit tech savy but he is no programmer and hasn't any idea of how complex work can get. He's been you know in an attempt to help me been setting out deadlines for project modules that are just so insanesly IMPOSSIBLE.

Like he thinks that by setting a deadline he can get the work done no matter how ridiculously insane those deadlines maybe. One module which he had laid out for me required that I:

    Integrate an ajax based webmail client i.e. roundcubemail

    Integrate tinymce rich text editor in the package.

    Add like 3 different plugins to the editor.

    Tweak the webmail client so it would send recieve requests upon sending a mail.

    Fit a tagging system into the mail client

    Add an automatic mail response function



And the time limit for all this ... 9 days and that too after I had to literally wrestle it from his original deadline of 5 days despite my asking for more time...

Note. This is all part of one module....and the number of programmers working on the project...including me just one..

I mean I can't take it any more - my boss is real nice but when it comes to setting deadlines he's killer!!!! ANd these deadlines are basically impossible - considering the fact that I had no prior knowledge of the workings of roundcubemail and tinymce 7 days ago - and even now I'm still learning how to tweak the packages . It isn't the kind of thing that can be done overnight and this is what my boss has a hard time UNDERSTANDING!!!

The projects aren't immediate deliverables but projects for our own company's use initially. I don't know how to talk to him on this. He just surfs through sites and when he sees somethings he likes. he just calls me saying:

OK I want that function from this site, a coupel of those from that site, this littel doo hickey form this site, a tagging system like they use in Gmail and wordpress in one package and I want it in two days....


And despite my begging that I need more time, he thinks I'm stalling or too lazy to work.. to be honest I haven't slept for a couple of days without the last time having to actually dream of function callbacks - I'm literally on the verge of explosion.

What do I do and how do I tell my boss - I've tried EVERYTHING? The stupid deadline is on the brink and I've somehow miraculously managed to finish about 75 percent of the tasks but in his case he's gonna be angry on the 25 percent that couldn't AND CAN NOT BE FINISHED in the time setup...

...Help me

(Edited by paritycheck on 05-02-2007 13:25)

Tyberius Prime
Maniac (V) Mad Scientist with Finglongers

From: Germany
Insane since: Sep 2001

IP logged posted posted 05-02-2007 13:58 Edit Quote

well - difficult situation, always, but a reasonable boss will listen if you provide enough documentation on what you did, and how long each part of your work took.

You might consider this being a good opportunity to say 'the buck stop's here - we're constantly missing deadlines. That means one of two things: a) I'm too slow (by what reason whatsoever) b) your deadlines are unrealistic. Now I need to ask you: Have I been pulling my weight? Are you making a profit on my work? Then it is b) and we need to rethink the way we set deadlines.

Honestly it sometimes pays to be frank.

hyperbole
Paranoid (IV) Inmate

From: Madison, Indiana
Insane since: Aug 2000

IP logged posted posted 05-02-2007 19:33 Edit Quote

Try reading Death March by Edward Yourdon. In fact any of the books listed in the above link would help you to survive this. Sometimes the best thing to do is to recognize that the only managment style your employer understands is a Death March and it's time for you to leave.

Telling the boss he is creating Death Marches and if he keeps it up you will have to find other options can sometimes shock them out of the behavior, but don't count on it. Also don't tell them you are ready to leave unless you really are. They may take you up on it. So, start preparing now by looking for other options before you approach your boss.

Look around to see what your other options are, then approach your boss for a frank an honest discussion about the situation. Be prepared to give in this discussion as well as take.

.



-- not necessarily stoned... just beautiful.

WarMage
Maniac (V) Mad Scientist

From: Rochester, New York, USA
Insane since: May 2000

IP logged posted posted 05-02-2007 22:01 Edit Quote

The advise above is great. I want to add a couple of things; You must be consistent, and documentation is your friend.

If you tell your supervisor it is going to take about 4 weeks, do not let him talk you down to 5 days. He can say I will give you 9 days to do it, and you would need to respond in a way that politely lets him know that you do not feel this is reasonable, and you still believe that it will take 4 weeks.

If he wants the task completed in 5 days you can ask him which parts of the project is he willing to live without. Tell him what would be reasonable to accomplish in 5 days.

When being asked for an estimate it is often a good idea to write a estimate report. You will want to describe each task in detail and then give reasonable numbers for each small step in the process.

For your list above you have -

* Integrate an ajax based webmail client i.e. roundcubemail
* Integrate tinymce rich text editor in the package.
* Add like 3 different plugins to the editor.
* Tweak the webmail client so it would send recieve requests upon sending a mail.
* Fit a tagging system into the mail client
* Add an automatic mail response function

You can turn the list above into -

code:
* Download all required software : 1 hour
* Research installation and configuration of the software : 8 hours
* Integrate roundcubemail into existing web application: 4 hours
* Integrate tinymce editor into roundcube mail: 4 hours
* Integrate 3rd Party Plugins: 2 hours (each)
* Code Inspection for tagging and auto-response integration: 16 hours
* Create custom email tagging system
    - Database schema updates for tagging: 4 hours
    - Application coding for tagging system
       * Add additional columns for tags display: 8 hours
       * Add administration section for tag maintainance: 16 hours
       * Add screens for tag CRUD: 16 hours
* Create custom auto-response system
   - Data base schema for response storage: 8 hours
   - Application coding for auto response: 16 hours

Total: 106 hours



When I write something like this up I normally also include a paragraph or two giving an overview of each task and what would be required.

With this kind of documentation you can ask your supervisor where he feels that you have overestimated. It is much easier to defend the time allocation to specific tasks than it is to defend an overall time on a nebulously defined project.

Dan
Code Town | Zombie Head | How Much TP?

(Edited by WarMage on 05-02-2007 22:01)

paritycheck
Bipolar (III) Inmate

From: you tell me
Insane since: Mar 2004

IP logged posted posted 05-03-2007 08:18 Edit Quote

That seems like very good practical advice... the problem being that even if somehow I do finish the work within the stipulated time [and that too after I give up on everything save a couple of trips to the bathroom and maybe food and a few hours of sleep surviving on just air and water ] - my boss then reduces further deadlines even further - he's acting like a slave rider in this respect infact I'm on the verge of bursting. It ain't easy being an only programmer.

Todays a meeting where my boss is going to set more deadlines, I'll use the chance to speak to him .. will let you guys know of what happens... wish me luck..

Tyberius Prime
Maniac (V) Mad Scientist with Finglongers

From: Germany
Insane since: Sep 2001

IP logged posted posted 05-03-2007 10:42 Edit Quote

yeah, hyperbole has a very good point as well - I pulled that once ( saying 'you're unreasonable - shape up, or I *will* walk out), and it was the right time, ergo the situation improved tremendously. But I would have walked if it hadn't.p

hyperbole
Paranoid (IV) Inmate

From: Madison, Indiana
Insane since: Aug 2000

IP logged posted posted 05-03-2007 18:14 Edit Quote

Dan,

An estimate report is a very good idea. I often use them to estimate the time a project will take.

I would add two things to what you have said:

1) Each step in the project needs to be divided into the smallest steps possible. The larger the step, the more you are guessing about your time.

For example, if a step in the project is "copy an xml file to the database". Divide this up into as many steps as possible, starting with
1. Read the xml file.
2 Parse the xml statements
3. store in the database.

Now break each of these into their component parts until you are at the point of (eg.) "use a library call to read a line of text", or "read a character from the open port"

Once you have reached this level of granularity in your analysis of the project, you can start assigning times to each task.

This is the where my second point comes in:

"No task takes less than half a day to complete."

This second point may seem a bit of an over estimate. You may look at a task and say to your self "I can write the function to read a character from the port in fifteen minutes." But, this doesn't allow any time for errors, testing, or unforeseen events. Also if it really only does take fifteen minutes to write that function, you have surely underestimated the length of time to write some other function so the extra time will be used up there.

I always take another step after coming up with the estimates for each process in the project: I will usually take the total time (for example Dan's 106 hours) add 10% and then double the time. This means that the above mentioned project will probably actually take about 230 hours to complete.

Again, this seems as if I'm being very liberal with the time, but I've been estimating projects for over twenty years and have many times received comments on how accurate my estimates are.

Of course, that may just be because, having done this for so long, I have a good feel for how long each project should take.

There was a fortune cookie on usb-4.2 that said "To make an accurate estimate of how long a project will take, make your best guess for the length of the project, add 10% and double the result. Then convert to the next higher units of measure." This means that Dan's estimate should be 230 days to complete the project.

.



-- not necessarily stoned... just beautiful.

poi
Paranoid (IV) Inmate

From: Norway
Insane since: Jun 2002

IP logged posted posted 05-03-2007 19:30 Edit Quote

Some friends came up with the PI rule. Estimate how fast you can do a task, then multiply this time by PI. This will cover the usual documentation lookup, hurdles, snafus, pebcak, and some QA.

Of course one could probably get the work done in the time estimated first by don't count on having something clean or maintainable.



(Edited by poi on 05-03-2007 19:32)

paritycheck
Bipolar (III) Inmate

From: you tell me
Insane since: Mar 2004

IP logged posted posted 05-04-2007 16:22 Edit Quote

I have to be one real sucker for sweet talk or my boss must be some manipulative genius. I did try to edge in some facts that the time wasn't enough keeping in mind that I was working with code I had not written, instead however my boss gives me a schoolyard type pep talk telling me to be be more confident in myself and stressed a lot on 'I know you can do it man, take it as a challenge - you don't know what you're capable of'. Well after a real long session which didn't turn out quite as I expected - I now have a deadline which on teh surface seems improbable and below simply insane.

I think I'll continue at my maximum sustinable pace i.e. the pace at which I can work without going clinically insane even if it means working overtime and see what I can do in the time - hey its not like I have a wife or kids to go home to so I can work late in the office until my boss realises that I'm overdoing it [that is if he ever does notice that ]. At the time of the deadline, he can't blame me for not working hard enough or long enough, it seems like a long term plan but anyway anything to get my mind of worrying about the deadlines...

What do you guys think?

Tyberius Prime
Maniac (V) Mad Scientist with Finglongers

From: Germany
Insane since: Sep 2001

IP logged posted posted 05-04-2007 16:41 Edit Quote

I think you are already out of your mind.
You'll simply burn yourself out for nothing much.
It does not pay to overwork knowledge workes - for your output will be just about the same wether you work
40 hours a week and relax, or you work 60-70 hours and leave the office just for sleep...

->Tyberius Prime

paritycheck
Bipolar (III) Inmate

From: you tell me
Insane since: Mar 2004

IP logged posted posted 05-04-2007 17:27 Edit