Way to Address Product Design Failure(core77.com)
core77.com
Way to Address Product Design Failure
https://www.core77.com/posts/138379/The-Best-Way-to-Address-Product-Design-Failure
10 comments
If I remember correctly, Peak Design, will replace/fix (if broken, do not work) a product throughout its lifetime. I remembered that one and was impressed for products that are likely to be torn, broken, with use.
Seriously impressive reply to a customer complaint. If only there was more of this in the world.
Something is wrong with this website. It gets stuck at 100% CPU and keeps allocating memory halfway through loading on FireFox. Archive version (https://archive.is/yfR6J) works fine.
The top banner also float in the middle when scrolling down or up.
When you go back from an article, the previous page reloads from the top and your position is lost.
It isn’t working that well on mobile, overall.
When you go back from an article, the previous page reloads from the top and your position is lost.
It isn’t working that well on mobile, overall.
Works fine for me in Firefox under Linux. (Maybe you're not running an ad blocker?)
I am using Firefox on Linux with almost all uBO filters enabled. Not sure what it could be
Must be an issue with the site itself when viewed in Firefox on Windows; I've run into the same thing every time someone shares it here on HN.
I even disabled uBO thinking that might be the cause, but the behavior didn't change. Eventually, I just gave up trying to access it.
Honestly, it's been disruptive enough that I'm considering adding the site to a personal blocklist, just to avoid running into it again. In the few times it's happened, the only way to recover was force-closing the browser, or waiting an excruciating amount of time and with a poorly responding browser try to close the tab.
Honestly, it's been disruptive enough that I'm considering adding the site to a personal blocklist, just to avoid running into it again. In the few times it's happened, the only way to recover was force-closing the browser, or waiting an excruciating amount of time and with a poorly responding browser try to close the tab.
BTW this approach, monitoring social media and responding personally, meshes well with the idea of doing things that don't scale. While you're still small, going out of your way to delight the customer is one of the stronger ways to build customer loyalty and brand awareness. (And your company may stay small for quite some time.)
As a software engineer, I was summoned twice from high-up to directly deal with an end-user issue.
Once, it was an issue with the installation of a legacy product I didn't work on that failed on a specific platform. I was handed a RDP session and had 30 minutes to figure it out. Somehow I pulled through, found the bug in the installer and fixed it live. Apparently the customer was at the end of their wits because they were bouncing back and forth with support for the past three months or so.
Another time, it was random crashes in a legacy product I didn't work on (again) that happened only with a specific combination of hardware and operating system. In three days, I've root caused it to the product's platform abstraction layer where the pthread mutexes were not created with the recursive flag. The bug was introduced over 10 years ago and the undefined behavior only triggered with a recent version of glibc running combined with a recent generation of Intel processors.
In both cases, I can't help but sympathize with the customer, who hit a real critical bug and had to escalate it all the way up until it reached me, the person you call when there is no one else left to call.
Once, it was an issue with the installation of a legacy product I didn't work on that failed on a specific platform. I was handed a RDP session and had 30 minutes to figure it out. Somehow I pulled through, found the bug in the installer and fixed it live. Apparently the customer was at the end of their wits because they were bouncing back and forth with support for the past three months or so.
Another time, it was random crashes in a legacy product I didn't work on (again) that happened only with a specific combination of hardware and operating system. In three days, I've root caused it to the product's platform abstraction layer where the pthread mutexes were not created with the recursive flag. The bug was introduced over 10 years ago and the undefined behavior only triggered with a recent version of glibc running combined with a recent generation of Intel processors.
In both cases, I can't help but sympathize with the customer, who hit a real critical bug and had to escalate it all the way up until it reached me, the person you call when there is no one else left to call.
I purchased a domain with Google Cloud a year ago for a business domain. There is no way to manage it. I get emails from SquareSpace noting they manage the domain. However, logging into SquareSpace they show no owned domains. Google Cloud says to go to SquareSpace. SquareSpace's AgentBot chats quickly give up and re-direct me to their FAQ, even after going thru a long process of uploading drivers licenses etc. The FAQ suggests I should contact the AgentBot. Meanwhile, the domain is up for rewnewal in 3months, so the problem has lasted for 9months no w/o resolution. Some online docs say Google Cloud domains got purchased by SquareSpace, others say only Google Domains got purchased by SquareSpace.
Meanwhile, not a single person responds. You cannot reach a human. You cannot contest it. I totally see the risk of going with Google for any other enterprise products.