# Thread: I don't need sanity, work in IT (thread) every day

1. Originally Posted by Shaikar
Today I had an estimate explained by a senior dev. Why would something take 120days?
Well, he'd estimated that the thing would take 6000 lines of code to write, based on the requirements.
From there an estimate is easy, as he knows he can write no more than 50 lines per day, so 6000/50 = 120 days.

Simples!

I find most of the time spent delivering good code is spent paring it down from the 60000 lines you wrote to the 6000 you actually need.

2. Originally Posted by Shaikar
Today I had an estimate explained by a senior dev. Why would something take 120days?
Well, he'd estimated that the thing would take 6000 lines of code to write, based on the requirements.
From there an estimate is easy, as he knows he can write no more than 50 lines per day, so 6000/50 = 120 days.

Simples!

It's very much dependent on circumstances. If you're talking about say frontend code then 50 lines per day is rather low, even if we include testing into the time needed. For this you basically write stuff down.
If on the other hand you're talking about a math heavy backend code, then 50 lines per day (average of course) is a rather heavy workload. Especially if this time includes testing. For this most of the time is spend on researching what the code should actually do as well as writing several times the code length in unit tests. Of course those unit tests also need braintime because they need to be good. And then there's optimizing the code.

So basically what I'm saying is that if the person is experienced (knows his average output) and knows his area of work, then estimating by total lines of code div lines per day is valid. Especially on super long estimates like 120 days, which nobody sane is expecting him to be perfectly accurate about.

3. Originally Posted by theBlind
So basically what I'm saying is that if the person is experienced (knows his average output) and knows his area of work, then estimating by total lines of code div lines per day is valid. Especially on super long estimates like 120 days, which nobody sane is expecting him to be perfectly accurate about.
Yeah, about that... My usual scenario with my boss is:
1. Give me an estimation
2. Can't you cut it down in half?
3. (nearing delivery date)You said it will be done in n time and it has to be done in n time.

4. Originally Posted by Corwyna
Originally Posted by theBlind
So basically what I'm saying is that if the person is experienced (knows his average output) and knows his area of work, then estimating by total lines of code div lines per day is valid. Especially on super long estimates like 120 days, which nobody sane is expecting him to be perfectly accurate about.
Yeah, about that... My usual scenario with my boss is:
1. Give me an estimation
2. Can't you cut it down in half?
3. (nearing delivery date)You said it will be done in n time and it has to be done in n time.
Which is why I added the "sane" as a disclaimer

5. Originally Posted by theBlind
Originally Posted by Shaikar
Today I had an estimate explained by a senior dev. Why would something take 120days?
Well, he'd estimated that the thing would take 6000 lines of code to write, based on the requirements.
From there an estimate is easy, as he knows he can write no more than 50 lines per day, so 6000/50 = 120 days.

Simples!

It's very much dependent on circumstances. If you're talking about say frontend code then 50 lines per day is rather low, even if we include testing into the time needed. For this you basically write stuff down.
If on the other hand you're talking about a math heavy backend code, then 50 lines per day (average of course) is a rather heavy workload. Especially if this time includes testing. For this most of the time is spend on researching what the code should actually do as well as writing several times the code length in unit tests. Of course those unit tests also need braintime because they need to be good. And then there's optimizing the code.

So basically what I'm saying is that if the person is experienced (knows his average output) and knows his area of work, then estimating by total lines of code div lines per day is valid. Especially on super long estimates like 120 days, which nobody sane is expecting him to be perfectly accurate about.
Yeah, totally this.

It can take far longer to write 50 lines of good code than 100 lines of bad code. Particularly if the 50 lines does the same thing as the 100 in a more elegant way.

6. applying hotfixes to 2008 R2 servers with massively backed up windows update problems to fix a file path length error so I can just run some fucking compare-object powershell scripts to find out if I can bin a load of stuff on their data partition \o/

Originally Posted by Liare
Originally Posted by Lana Torrin
Originally Posted by Liare
Originally Posted by Lana Torrin
So news broke on the long weekend that our sdm had finished the image.. Want to place bets on how fucked it is?

Also everyone keeps insisting on using capture images in sccm.. I don't get it, but because I'm literally the only person to not have sccm training (and the only person to actually do a full windows 7 deployment with an install image) I just get ignored..
usually you pre-bake the image to make the rollout less annoying, update rollups and such.
You can do all that with an install though, plus it makes patching it easier because sccm knows how to inject the patches, you don't have the hassle of conflicting drivers because it deploys them based on rules... It's faster to build. Literally the only down side is no baked in apps but if you have them all packaged anyway (which you should) it'll just deploy them after setup..

Everyone seems to be stuck in a 'this is how you did it in xp so we aren't changing' mindset.
you're forgetting a important factor here, i dont blame you for forgetting it, but you're forgetting it.

SCCM is shit, i worked with it professionally the last six years and i can say it's the bit of the job i miss the least, i missit even les than i miss the awful user interactions, that's how shit it is.
Wait I'm going in for training on this exact shit soon /o\

7. Originally Posted by Frug
Originally Posted by theBlind
Originally Posted by Shaikar
Today I had an estimate explained by a senior dev. Why would something take 120days?
Well, he'd estimated that the thing would take 6000 lines of code to write, based on the requirements.
From there an estimate is easy, as he knows he can write no more than 50 lines per day, so 6000/50 = 120 days.

Simples!

It's very much dependent on circumstances. If you're talking about say frontend code then 50 lines per day is rather low, even if we include testing into the time needed. For this you basically write stuff down.
If on the other hand you're talking about a math heavy backend code, then 50 lines per day (average of course) is a rather heavy workload. Especially if this time includes testing. For this most of the time is spend on researching what the code should actually do as well as writing several times the code length in unit tests. Of course those unit tests also need braintime because they need to be good. And then there's optimizing the code.

So basically what I'm saying is that if the person is experienced (knows his average output) and knows his area of work, then estimating by total lines of code div lines per day is valid. Especially on super long estimates like 120 days, which nobody sane is expecting him to be perfectly accurate about.
Yeah, totally this.

It can take far longer to write 50 lines of good code than 100 lines of bad code. Particularly if the 50 lines does the same thing as the 100 in a more elegant way.
I really try to hammer this home to young guys. They will not be judged on the number of lines they write. Quite the opposite. 10 lines of great, idiomatic code are worth a whole lot more to us than a bunch of busy work. I would also rather have them false start several times and throw away those learning attempts than clinging to bad ideas. I find the old adage, 3rd times the charm to be very true in this regard.

8. Agreed.

As soon as you actually start writing code, all the (hard) work is essentially done. What remains is a simple exercise in typing.

9. Mates I'm pretty sure I've just traced a compromised log in to a VPS in croatia wtf

10. It wasn't me.

11. My hosting company just informed me- in no uncertain words, that one of my (private) servers is part of a DDoS attack and that they will shut it down (and keep it that way), if I don't swiftly remedy that situation.

Turns out I had my DNS server configured as an open relay (don't ask, I have no idea)... I blame, well ... me

12. Originally Posted by Hel OWeen
My hosting company just informed me- in no uncertain words, that one of my (private) servers is part of a DDoS attack and that they will shut it down (and keep it that way), if I don't swiftly remedy that situation.

Turns out I had my DNS server configured as an open relay (don't ask, I have no idea)... I blame, well ... me
That is similar to how I just gave up on Linode after half a decade hosting my rtorrent project there.

My mailing list's email server or something got exploited and used as a relay, then the Linode people decided to contact me by sending me a private message that I could only see if I logged on to their website. Not a single email sent, and since I didn't reply for weeks since why the fuck would I know, they cut my node from the net and decided to be assholes about what I needed to do to get it back.

So yeah, assume your customer is psychic and then blame them for not fixing an issue that they obviously should have known about because the customer should be psychic.

13. Originally Posted by Rakshasa The Cat
Originally Posted by Hel OWeen
My hosting company just informed me- in no uncertain words, that one of my (private) servers is part of a DDoS attack and that they will shut it down (and keep it that way), if I don't swiftly remedy that situation.

Turns out I had my DNS server configured as an open relay (don't ask, I have no idea)... I blame, well ... me
That is similar to how I just gave up on Linode after half a decade hosting my rtorrent project there.

My mailing list's email server or something got exploited and used as a relay, then the Linode people decided to contact me by sending me a private message that I could only see if I logged on to their website. Not a single email sent, and since I didn't reply for weeks since why the fuck would I know, they cut my node from the net and decided to be assholes about what I needed to do to get it back.

So yeah, assume your customer is psychic and then blame them for not fixing an issue that they obviously should have known about because the customer should be psychic.
That's kind of fucky. They have the ability to send you a PM, but that doesn't translate to an actual email? Won't be using them in a hurry.

14. Originally Posted by erichkknaar
That's kind of fucky. They have the ability to send you a PM, but that doesn't translate to an actual email? Won't be using them in a hurry.
Yeah, they used to be really good since the start and then suddenly they went full retard.

When I finally realized what was happening I replied and told them I'll fix the issue just give me access again... which was late evening Japanese time... and the next day while at work I tried to deal with it turns out they'd already cut the node off again. Kept an open relay node running for several weeks without sending an email yet close it off after a few hours while the customer is sleeping, then blame the customer again.

Decided not to waste time and switched to AWS.

15. Originally Posted by Rakshasa The Cat
Originally Posted by erichkknaar
That's kind of fucky. They have the ability to send you a PM, but that doesn't translate to an actual email? Won't be using them in a hurry.
Yeah, they used to be really good since the start and then suddenly they went full retard.

When I finally realized what was happening I replied and told them I'll fix the issue just give me access again... which was late evening Japanese time... and the next day while at work I tried to deal with it turns out they'd already cut the node off again. Kept an open relay node running for several weeks without sending an email yet close it off after a few hours while the customer is sleeping, then blame the customer again.

Decided not to waste time and switched to AWS.
Yeah, I'm primarily GCE these days. AWS storage is slightly cheaper, but I find GCE better for actual instance time and network. Much more developer friendly tool set as well.

16. Win10 pro image has been deployed on to 1 desktop and 1 laptop. We have been told we have 2 days to look at it before it goes to a pilot group of 20 people with full deployment starting next week.

Project manager was promptly told we aren't looking at dick without time on the project because we all have other work we are supposed to be doing.

I also noticed they put all the machines in my test OU that has inheritance blocked (and a GPO that's not going to work), so this is going to go well.

17. Sometimes you just have to put all of your masters in a room together and see which one wins.

18. Project manager is rolling out to test group Monday despite being told the new win10 boxes don't map the drives properly (or at all). Fun will be had I'm sure.

19. There's a MS patch from last year that prevents GPOs for mapped drives being applied on clients due to how the GPO is created. A similar thing happens with printer mappings. Maybe that's the problem?

https://support.microsoft.com/en-us/...-june-14,-2016

tl;dr
Symptoms
All user Group Policy, including those that have been security filtered on user accounts or security groups, or both, may fail to apply on domain joined computers.

Cause
This issue may occur if the Group Policy Object is missing the Read permissions for the Authenticated Users group or if you are using security filtering and are missing Read permissions for the domain computers group.

Resolution

To resolve this issue, use the Group Policy Management Console (GPMC.MSC) and follow one of the following steps:

- Add the Authenticated Users group with Read Permissions on the Group Policy Object (GPO).
- If you are using security filtering, add the Domain Computers group with read permission.

20. Already know about that one but that's not it.. It would be the GPO blocking mentioned above, but for some reason they seem to think it's the AV. I'm not going to mention anything because I hate this client and also the project manager, so they can waste time chasing nothing.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•