Repository navigation
[Due for payment 2026-10-02] Payment issue for [Commuter Mileage Exclusion] [R1] regression consolidation - @marufsharifi #100085
Description
Claude posting on behalf of Matt:
Payment thread for the Contributor+ work on the [Commuter Mileage Exclusion] [R1] project, reviewed by @marufsharifi on #93848.
The regression issues that spawned from that PR are being closed with no payment. The total for the project gets discussed and settled here, once everything has been on production for a week. @Gonals agreed to this approach on #98248.
Why this issue exists. The project's parent issue lives in the Expensify/Expensify repo, which external contributors can't access. Pointing the payment at a thread @marufsharifi can't open doesn't work, so the settlement happens here instead.
Closed with no payment
- [Due for payment 2026-08-21] Distance - Manual and Odometer tab are shown when tracking distance from expense report #98225
- [Due for payment 2026-08-21] Distance - Manual and Odometer tab are shown when every workspace has exclusions #98248
- [Due for payment 2026-08-21] workspace distance settings link is clickable for workspace member and it opens not here page #98252
- [Due for payment 2026-09-09] Expense-App crashes on creating workspace after creating distance expense. #98365
Open, flagged under the same rule
- [Due for payment 2026-08-21] Distance - Manual distance can be moved to workspace with fixed distance exclusions #98222
- [Due for payment 2026-09-18] Distance - Distance can be edited via command when exclusion is enforced #98223
- [Due for payment 2026-09-09] Android - Distance - "Map or GPS distance required" overlaps with the image #98228
- [Due for payment 2026-09-18] Distance - Original and Removed distance shows wrong amount after changing unit and rate #98231
- [Due for payment 2026-09-02] Distance-Stale Original and Removed commuter distance are shown after editing distance offline #98251
- [Due for payment 2026-09-02] Distance - Original and Removed commuter distance shows new exclusion instead of edited distance #98257
- [Due for payment 2026-09-04] Distance - Distance field shows wrong expense when merging distance expense with exclusion #98343
Already paid
- [Due for payment 2026-08-21] Distance - Missing Original amount and Removed commuter miles when submitted from self DM #98229, $250 on 2026-08-26. That predates this approach and comes off whatever total we land on.
Nothing is owed on the individual issues above. Post questions about the amount here.
Activity
melvin-bot commentedon Sep 1, 2026
@mallenexpensify — to hire @marufsharifi on Upwork, add the External label so a job is created and the offer is sent automatically.
mallenexpensify commentedon Sep 3, 2026
Testing summin' with the
Needs Reproductionlabel.melvin-bot commentedon Sep 5, 2026
@marufsharifi Whoops! This issue is 2 days overdue. Let's get this updated quick!
melvin-bot commentedon Sep 7, 2026
@marufsharifi Eep! 4 days overdue now. Issues have feelings too...
melvin-bot commentedon Sep 9, 2026
@marufsharifi 6 days overdue. This is scarier than being forced to listen to Vogon poetry!
mallenexpensify commentedon Sep 10, 2026
@marufsharifi I made myself owner and made this weekly. Ping me when all PRs have been on production for 7 days and we can discuss payment.
16 remaining items
mallenexpensify commentedon Oct 8, 2026
Claude posting on behalf of Matt. This is a test of a new review process. The rubric behind proposed amounts is not public yet. Reply if a number looks wrong.
This issue has 12 merged PRs. Above four PRs, the review classifies each PR and proposes no total. @Gonals sets the amount.
Unpaid, 10 PRs. #98281, #98368, #98287, #98571, #98321, #98371, #98273, #98286, #98427 and #99657 fix regressions from #93848 or #98321. @marufsharifi reviewed both of those PRs. CONTRIBUTING.md states: "Work to fix a regression caused by your PR is unpaid by default."
Due, 1 PR. #98423 fixed #98365. @fedirjh traced #98365 to #93915, and @marufsharifi is not on #93915. Full job price, $250.
Needs a ruling, 1 PR. #100478 fixed #99710 and #99713. Those came from #94316, which @marufsharifi did not review. They also involve the #98281 guard, which he did review.
Original review of #93848. The 70 files support an increase. The 11 regressions support a cut. Both apply.
Already paid: $250 on #98467 and $250 on #98229.
Claude posting on behalf of Matt.
PAYRUBRICTEST v1mallenexpensify commentedon Oct 8, 2026
Claude posting on behalf of Matt. This is a test of a new review process. The rubric behind proposed amounts is not public yet. Reply if a number looks wrong.
@mallenexpensify added the below after the post cuz the review process is currently built for bugs
$125 outstanding to @marufsharifi. @Gonals for approval.
Two rules for this project:
$0, 10 PRs. Each fixes a regression from a PR that @marufsharifi approved:
$250, #98423. Melvin named #93848 for #98365, but @fedirjh traced it to #93915. @marufsharifi is not on #93915. 10 files, standard size.
$125, #100478. #99710 came from #94316, which @marufsharifi did not review. #99713 came from #98281, which he approved. The PR changes 21 lines outside tests. 50%.
Original review of #93848. The $250 paid on #98467 stands. The 70 files support an increase, and the 11 regressions support a cut. This review proposes no change.
Math. $250 + $125 = $375 for the fix PRs. #98229 already paid $250 for #98321, which is a $0 PR under rule 1. $375 - $250 = $125 outstanding.
Rubric rules used: "0%: the regression came from the payee's own PR", "The named offending PR is a claim", "Every amount is per payee".
Claude posting on behalf of Matt.
PAYRUBRICTEST v1mallenexpensify commentedon Oct 8, 2026
Added this above
melvin-bot commentedon Oct 12, 2026
@Gonals Whoops! This issue is 2 days overdue. Let's get this updated quick!
mallenexpensify commentedon Oct 12, 2026
@marufsharifi can you review your increase request, taking into account that, by default, we don't want to pay $250 for PRs where you introduced the regression? @Gonals can you review once @marufsharifi posts? Thx
Gonals commentedon Oct 13, 2026
Will do
marufsharifi commentedon Oct 13, 2026
@mallenexpensify, I understand that regression fixes are unpaid. This was a whole new feature, though, and some follow-up work addressed behavior that needed further clarification, so I don’t think all of it should automatically be treated as regression fixes. I’ve shared the full context from my side here: #100085 (comment), and would appreciate @Gonals deciding what’s fair for the overall feature review. I’m happy to accept his decision. Thanks.
mallenexpensify commentedon Oct 13, 2026
@Gonals please review again. With you leading the new feature, it's your responsibility to either set an amount beforehand or to review after to help determine a fair price, the former is best practice to help avoid confusion later (which can still happen)
Gonals commentedon Oct 15, 2026
For new features, I'd say we consider it regressions (so not subject to additional payment) if they should have been fixed/caught in the original PR.
In this case, there are a lot of moving parts and it is within expectations that some things may not be caught in the original issue.
Reviewing the list, I'd considere regressions (should/could have been caught while working on the original issue) the following:
#98365
It caused a crash, so it is hard to argue it is not a regression (even if it was safe behind a beta).
#98225
#98248
Manual/odometer hiding was specified in the design, so it should have been handled properly.
#98251
#98257
Editing is core functionality and should have been handled properly
If we consider those regressions, we should deduct $625 from the total, so the remaining payment should be $1500