Dana

Cost is worth stating plainly since you'd be paying again. The well-known subscription ones sit around $99 to $109 a year now: one of them is $109 annually or $14.99 monthly, and it lets you share the subscription with up to six people, which matters if there's a household involved.

Against twenty minutes a week, that's roughly $2 an hour of your time back. That's either obviously worth it or obviously not depending on whether the app fixes the skipping, which no subscription can promise.

You gave up heavy hinging and easy leg loading. Everything else you can do at home better. Add one loaded hinge movement a week and the trade is basically neutral.

The one sentence rule is brutal and correct. About eight of mine fail it immediately.

Sleeping nap is a fourth one, from Old English hnappian, to doze. So the score is four independent naps, which is a decent haul for one syllable.

Wallpaper. Sounds daft, works instantly. Solid dark blue on one, solid rust on the other, and set the browser and terminal themes to match. Peripheral vision catches colour before it catches window titles.

Going to disagree with the assumption that this is a fight. I asked three former clients directly whether they minded me selling a generic version of the tooling, plainly, over email.

All three said yes without hesitation. Two asked for a discount if they ever moved onto the product, which I gave. One asked me to confirm none of their data or business rules went with it, which was reasonable and true.

People assume the client will object and never test the assumption. A client who hired you for three days of work does not think of themselves as owning a software company.

Also worth noticing that the sticky notes are working. That's not fraud, that's data. They work because there are six of them, they're visible without an action, and they get thrown away. Whatever you build next should have those three properties.

Two to three months for most people I've talked to. It never went fully quiet for me, but it dropped from a hum to something I notice maybe once a week.

Agreed on the tax code point specifically. Mine was wrong for four months on the second job and I was overpaying the whole time without noticing, which took a phone call and a lot of hold music to sort out.

One more that bites people once: Secure cookies require HTTPS, and if your staging environment terminates TLS at a load balancer and talks plain HTTP to the app, some frameworks decide the request is insecure and silently refuse to set the cookie. Look for a trust proxy setting.

Worth adding that this also means your home IP is not in anyone's logs, which for a public facing hobby app is a real benefit.

On the subscription question specifically: at least one of the mainstream ones lets you share a single subscription across up to six people, and it's $109 a year or $14.99 a month, so a couple is paying one fee rather than two. That was the deciding factor for us over paying twice for parallel accounts.

Worth checking the sharing terms before you buy rather than after, the pricing pages are clearer about this than the marketing pages.

Cronometer's free tier is what I settled on and the trade off is clear: the food database is smaller and more curated, so fewer junk crowdsourced entries but more manual adding for own brand products, and the micronutrient detail is far better than anything else free. If your priority is that supermarket own brands scan reliably in the UK, look at anything built on Open Food Facts, since that database is open and community maintained and its coverage of European retail products is much better than the American focused apps. I ran both for a fortnight and kept the one with fewer wrong entries rather than the one with more entries.

Worth adding the distinction that trips up half the people asking this, because there are two different credits and they behave differently. The residential clean energy credit, the solar and battery one, is nonrefundable but carries forward. The energy efficient home improvement credit, the one for heat pumps, insulation and windows, is also nonrefundable but the IRS says explicitly that you cannot apply any excess to future tax years. So heat pump people who assumed they had a carryforward do not have one, and it simply evaporates.

You answered your own question in the middle of the post. Eating at maintenance is a decision to keep your current amount of tissue, and it works - you kept it.

Strength went up because your nervous system got better at the movements, which is most of what happens in the first year regardless of training style. Size needs a surplus, and it doesn't need a big one. This isn't a bodyweight-versus-weights thing at all; a barbell trainee at maintenance would be sitting here with the same tape measure.

Then over-deliver on it for the first month. Post before 9:00 every day, be specific, flag blockers early. After four weeks of that nobody remembers you're not on the call.

Ask one of them to let you go async, and ask for it as a working style thing rather than a scheduling thing. The phrasing that has worked for me three times: I get my best focused work done first thing, I'd like to post my update in the channel by 9:00 each day instead of attending, and I'll join live on Mondays and any day there's something to discuss.

Most managers say yes to that because a written update is genuinely more useful to them than watching someone recite it. Pick whichever team already communicates in writing a lot, that's the one where it'll land.

Simple test I use: if fewer than about a third of customers would use it, it is an add-on. If more than about two thirds would, it is a price rise. In between, ship it to everyone and raise new prices, because the packaging complexity is not worth the marginal revenue.

Longer term suggestion: reverse proxy the API onto the same origin as the frontend and CORS stops being part of your architecture. Every project I've moved to that setup has permanently deleted a category of bug from its life.

I'd argue against automating them yet, and I know that's the unpopular position at 30 customers.

Those calls are the only place you watch someone use your product without a filter. I killed mine at about 40 customers because I was busy, and my roadmap got noticeably worse within a quarter - I started building things people asked for in emails, which is a much more biased sample than watching someone get stuck in silence.

What I'd do instead: cut the call to 20 minutes by pre-doing the setup yourself before they join, and keep the watching part. Automate the clicking, not the conversation.

The bit people find genuinely surprising: a cheque can be returned unpaid weeks later in some systems, long after the funds became available. Availability is a promise about when you can spend it, not a guarantee that it's yours forever.

Politely disagreeing with the implied "paid equals private." Paying changes the incentive, it doesn't change the policy, and the policy is the only thing that binds them.

Read the section about aggregated or de-identified data specifically. Plenty of paid products still reserve the right to sell anonymised spending trends, which is a real market. If a paid app's policy is silent on that, that's the same silence a free one gives you, just more expensively.

Watch what it never says. That's the fastest read on how any free product pays for itself.

Going to argue the unpopular one. For five pages and a blog updated twice a year, WordPress is a maintenance liability you have volunteered to own. It needs core updates, plugin updates, a security posture and a backup regime, forever, for a site that changes twice a year. A static site with a small hosted editor, or even a site builder the client can log into, costs less to run and cannot be compromised through a plugin.

If the client insists on WordPress because they have heard of it, that is a real reason and you should charge for it. But do not pretend it is the cheap option.

Disagreeing with gating collaboration and export, which is where a lot of people reach first. The people who invite colleagues and pull data into their own reports are the ones who make you spread inside a company, and charging them for it slows down the mechanism that grows the account. Charge for volume, charge for the corporate requirements, and let the evangelism features be free.