Marley Spoon vs Hello Fresh: an update

A couple of months ago, after my second week of trying Marley Spoon, I wrote a post summarising what I’d found using it and Hello Fresh.

You might think this is an update to tell you about the third and subsequent weeks – after all, I did say at the end of that post that I’d be sticking with it but keeping HelloFresh around to try out occasionally.

Things didn’t work out that way though. The first week with Marley Spoon, my food had been left on the street, in the rain, at 2am – 4 hours before the 6am delivery window started. I updated my delivery instructions to explicitly tell them to call me at any time of day or night when the box was delivered; but despite this, the second week’s delivery was again delivered in the middle of the night and I wasn’t called.

The third week my box didn’t turn up at all. One of Marley Spoon’s features is that you can choose which 3 (out of 7) recipes you want in your box each week. Apparently when I made changes to my delivery the website decided that I really wanted to cancel the delivery. I wasn’t notified of this until after my box didn’t arrive.

So I switched back to HelloFresh and I’ve been using them every since. Their boxes arrive on time, and their recipes don’t have missing steps or missing ingredients.

I’d like to give HelloFresh a try again at some point in the future, but it won’t happen when their facebook page has posts likethis or this or this sitting on top. It sounds like they still haven’t got deliveries sorted out – which is a bit of a shortcoming in a delivery service.

Marley Spoon vs Hello Fresh

After a month or so of HelloFresh​, I’ve just finished my first week of Marley Spoon​.

I’m not going to go into detail – Techly’s review does a good job and matches my experience.

Two exceptions:

Firstly – Techly’s review mentions that Marley Spoon has limited delivery hours. I picked the 6-9am delivery slot, because I wanted to make sure I was home and the box didn’t end up sitting on the street for hours – but it ended up being delivered at 2:35am and sitting on the street (in the rain!) for hours. MS have updated their website to advise that the delivery window is now 2-8am, and I’ve asked them to call me whatever time of night they leave the box – we’ll see how that goes tomorrow.

Secondly: Techly’s review describes MS’s recipe cards as “easily the best to follow”. I haven’t found that to be the case.

Partly this is because HelloFresh has a really nice iOS app with very clear step-by-step instructions (including timers for any of the steps that need to be timed), while Marley Spoon only has the paper instructions, and they are often a jumble. They have to fit everything into 6 steps, so one single step might have several unrelated actions in it – “Bring some water to the boil. Meanwhile, fry the fennel”. They overlook kitchen implements (the recipe I cooked last night needs a stick mixer, but this isn’t mentioned in the list of implements) and ingredients (a stir fry came with oyster sauce and five spice powder, but the recipe didn’t mention using them – I had to make my own guess about when to add them).

Still, on the whole, I think I enjoyed Marley Spoon a little more. The recipes are a little bit more of a challenge, which I like (HelloFresh is probably better for beginners) and more varied. HelloFresh is still good though, and I’m keeping my subscription active (but on pause) for now – I plan to swap between the two every now and then.

Both services allow you to opt-out of individual weeks, so I’m signed up to both at the moment but can choose which I get each week. Marley Spoon offers your choice of 3 (or more, if you pay for more) recipes out of their selection of 7 each week; while HelloFresh gives you 3 fixed choices.

BTW, if you want to try out HelloFresh, use the referral code BW2MLZ – you get $35 off your first box, and I get $20 off my next box. Marley Spoon don’t seem to offer any kind of referral program.

Reproduction

(If this post doesn’t make sense to you, you haven’t been following the discussion on IRC that this relates to – feel free to ignore and go about your day)

Setup

  • virtualenv repro
  • cd repro
  • source bin/activate
  • wget https://jamezpolley.com/wp-content/uploads/2015/04/repro.tar.gz
  • tar -zxvf repro.tar.gz
  • pip install -f index/ bar==2.0

Installing foo directly

Expected behaviour
Foo and its dependency, bar==2.1, get installed
Observed behaviour
As expected

⌂64% [:~/src/tmp/repro] [repro] master* ± pip install -f index/ foo
Collecting foo
Collecting bar>=2.1 (from foo)
Installing collected packages: bar, foo
Found existing installation: bar 2.0
Uninstalling bar-2.0:
Successfully uninstalled bar-2.0
Running setup.py install for bar
Running setup.py install for foo
Successfully installed bar-2.1 foo-0.3
[:~/src/tmp/repro] [repro] master* ±

Cleanup

  • pip uninstall foo
  • pip install -f index/ bar==2.0

Installing requirements

Expected behaviour
Foo and its dependency, bar==2.1, get installed
Observed behaviour
bar==2.0 is left installed

⌂69% [:~/src/tmp/repro] [repro] master* ± pip freeze
bar==2.0
⌂68% [:~/src/tmp/repro] [repro] master* ± pip install -f index/ -r requirements.txt
Collecting foo (from -r requirements.txt (line 1))
Requirement already satisfied (use --upgrade to upgrade): bar>=2.0 in ./lib/python2.7/site-packages (from -r requirements.txt (line 2))
Installing collected packages: foo
Running setup.py install for foo
Successfully installed foo-0.3
⌂66% [:~/src/tmp/repro] [repro] master* ±

iOS “append to day note” workflow

I’ve created a Workflow which asks for input and appends it to a day note in Evernote.

For my reference, when I need to restore it onto another device, it’s available at:

https://workflow.is/workflows/45eabe23d0a04fadb48b62ec92147b3b

My Todoist homepage

This is what I use as my Todoist home screen:

od, @inprogress, (@today | today) & !@inprogress , @next, (7 days & !today &!@TODAY &!@inprogress)

Anything that shows up in “overdue” is immediately acted upon: cancelled (if there’s no point now that the deadline has passed) or completed (if I did do it but forgot to note the fact); rescheduled for today, if I can commit to getting it done today, or rescheduled for later.

Anything that I’m in the middle of working on, but have had to step away from, is labelled as INPROGRESS to remind me to finish those tasks first.

The next block is a mix of things scheduled for today (mostly recurring tasks) and stuff I’ve committed to getting done today (by labelling with TODAY

mtr on mavericks via rudix(‘s ports collection)

With thanks to https://code.google.com/p/rudix/wiki/HowToMakeNewPackages and https://trac.macports.org/changeset/112285

hg clone https://rudix.googlecode.com/hg/ rudix
cd rudix/Ports/mtr

You’ll need to update the Makefile:

diff -r 11e7e05f8dd7 Ports/mtr/Makefile
--- a/Ports/mtr/Makefile Fri Jul 20 18:29:04 2012 -0300
+++ b/Ports/mtr/Makefile Thu Feb 13 16:03:51 2014 +1100
@@ -2,7 +2,7 @@

Title= MTR
Name= mtr
-Version= 0.82
+Version= 0.85
Revision= 1
URL= ftp://ftp.bitwizard.nl/mtr/
Source= $(Name)-$(Version).tar.gz

Now you’ll need to provide one small patch to be applied while building:


mkdir patches

Make yourself a patch file that looks like this:

diff -rupN mtr-build.orig/asn.c mtr-build/asn.c
--- asn.c 2014-02-13 15:59:36.000000000 +1100
+++ asn.c 2014-02-13 15:59:46.000000000 +1100
@@ -21,7 +21,7 @@
 #include
 #include <sys/types.h>

-#ifndef __APPLE__
+#ifdef __APPLE__
 #define BIND_8_COMPAT
 #endif
 #include <arpa/nameser.h>

now you can

make build

cd mtr-build/
sudo make install

Money tracking/budgeting apps

A mini-review…

I’ve been using PocketMoney from Catamount Software on and off for a long time. It’s very powerful – but I always found that, unless I had a particularly urgent need for budgeting and tracking my spending, I was never motivated to keep up the effort. Every other program I’ve tried, including the ones I’m about to mention, feels like I’ve traded away some subset of PocketMoney’s power.

Currently I’m using two different apps (yes, that does mean that I’m entering every transaction twice. I’m not sure how long I’m going to keep doing this). One app is fantastic for forecasting your cash flow down to the last cent and second, so that you know exactly how much money you’ll have on hand at any given time.

The other app explicitly refuse to enable that kind of forecasting. Instead, it takes the approach that you should only ever make a budget for the money you already have on hand. If you have long-term goals, you can start to meet those goals by setting aside some of the money you have on hand, but you should never start playing with money you don’t actually have yet.

The first app – the one that does the forecasting is MoneyWiz

It’s fairly simple to use, but has almost all the functionality you’d expect to be able to track transactions on your accounts. Its best feature is the ability to set up recurring (eg, $x salary in to account 1 every fortnight, $y for a recurring doctor’s visit every 3 weeks) and one-off (eg, need to buy a birthday present for mum on june 15) scheduled transactions and then generate a forecast. For instance, I’m in the middle of planning to move house – I used this just last night to check how different levels of rent would affect me, taking into account all the other bills and living expenses I already know about.

The iphone, ipad, and Mac versions of MoneyWiz all support all the features and they all sync together, but you have to pay for each version separately (5.99 each for iphone and ipad, 24.99 for Mac). It’s possible to just get the iPhone version and do everything with that, but I find it handy to use the ipad or desktop version to be able to see more detail, especially when I’m looking at the forecast graphs.

The other one I’m trying out is You Need A Budget, or YNAB for short. It comes in Windows and Mac desktop versions, and has iPhone and Android mobile clients to support the desktop program. The mobile clients are free, but they only support a limited range of functions. You need to have paid for the desktop version to use them, and they’re mostly designed for you to check your budget balances or enter a transaction. YNAB has a free 34 day trial, but after that it costs $60.

YNAB doesn’t do forecasting; instead it focuses on having you figure out how much money you have *right now* and what you need that money to do before your next pay. They don’t explicitly say this, but obviously they believe that once you’re collecting the data about what you *have* been spending money on, you’ll be more able to make decisions that support your goals in the future.

As well as giving you 34 days to try it out, they have a comprehensive series of videos (http://www.youneedabudget.com/support – look for “YNAB 4 Video Tutorials” on the right) and tutorials (http://www.youneedabudget.com/support/category/the-ynab-method) which don’t just cover how to use the program, they aim to teach you how to think about budgeting.

If you’re looking for help budgeting because you just don’t know where the money goes, I recommend giving YNAB a go – after all, it’s free to try for 34 days (and there’s your first savings goal right there – have $60 on hand to pay for it at the end of those 34 days!). Take the time to watch some of their videos and sign up for their 9-day email course – Even if you don’t end up paying for it, the information you’ll get from the progam about your own spending, and from their courses about what to do next, will leave you much better able to make informed decisions about your spending.

If you have some idea of where your money goes but need better visibility into some of the trickier patches, MoneyWiz is probably a better choice. YNAB is aimed at helping you spend the money you have; MoneyWiz is much better at helping you plan what to do with the money you’re going to get later.

Personally, I’ve always been *horrible* with money. It comes in, at some point later it runs out, and then there’s a bit of a panic until the next payday. MoneyWiz has helped me a lot: because I was able to see ahead, I started having the panic weeks before the money ran out – which meant that the money ended up *not* running out. However, that’s turned into a steady stream of just-scraped-through paydays. I don’t end up in the same panic any more, but I do end up only *just* scraping through.

I’ve only been using YNAB for a few days, but the way it presents essentially the same information as MoneyWiz in a slightly different format, combined with their propaganda, have already made a difference in the way I’m thinking about budgeting.

I think I might need to keep using MoneyWiz for a few weeks (maybe even months) more to help me past one last tricky place – but as soon as I have a bit of a buffer (nb: thanks to MoneyWiz’s excellent forecasts, I know to the day when that’s going to be) I won’t need to MoneyWiz’s level of precision, and I’ll probably switch to just using YNAB.

Netcat tarpipe – with bonus progress bars!

This post was written for my own reference, so I can stop recreating the process each of the 2-3 times a year I need to use it

Every now and then, you have a large number of files to transfer across a really fast network, and the usual methods just have too much overhead.

At times like this, Ye Olde Skool Neckbeard Sysadmin reverts to a time-honoured technique known as the Netcat Tar Pipe:


On the receiving end do:
# netcat -l -p 7000 | tar x

And on the sending end do:
# tar cf - * | netcat otherhost 7000

The chiefest drawback of this technique is that you don’t know what’s happening. You know that Some Data is being transferred at A Rate, but that’s about all.

Enter Pipe Viewer, aka pv:


pv – Pipe Viewer – is a terminal-based tool for monitoring the progress of data through a pipeline. It can be inserted into any normal pipeline between two processes to give a visual indication of how quickly data is passing through, how long it has taken, how near to completion it is, and an estimate of how long it will be until completion.

A minor tweak or two will give you fancypants progress indicators, and a file transfer mechanism that’s almost certainly faster than anything else you can do.

First you need to calculate the size of the data you’re sending:


[/share/polleyj] # du -sk
594409480 .

Now we use a simple modification of the command above on the receiving side:


netcat -l -p 7000 | pv -s 594409480k | tar vx

and on the sending side:


tar cf - * | pv -s 594409480k | netcat otherhost 7000

You now get a fancy progress indicator on the sending side:


959MB 0:03:18 [4.64MB/s] [> ] 0% ETA 33:13:22

and with the addition of the ‘v flag on the receiving side, you can see filenames as they’re unpacked as well:

Bodacious/Downloads/._.DS_Store
Bodacious/Downloads/VERY_LARGE_FILE.ISO
352MB 0:02:41 [4.85MB/s] [> ] 0% ETA 73:41:36

So there you go. May as well go make a coffee or 73 while you wait for that data to copy over.

Top tips from #pyconau

Last weekend I was at PyCon-AU in Hobart. Plenty has been said, on twitter and else where about what a great conf it was, so I won’t go into that too much.

I will mention that my biggest complaint is that there were too many talks that I wanted to see, so I missed about 2/3rds of them simply through being unable to be in more than one place at once. Fortunately all the talks that I missed are available on YouTube so I’ll be gradually catching up on them as time permits.

I came away from the conference with, amongst other things, a new grab-bag of tools that I plan to be using shortly. Some of the most valuable are:

I’m already excited about next year. Terrifyingly, I’ve already started planning a couple of talks I’m going to propose.

Wrest Point Casino hotel swimming pool length

I’ve been unable to find this information online, so I thought I’d fix the problem.

The hotel pool at Wrest Point Casino is roughly 10m long. For the convenience of patrons, it has handrails at each end, just underwater. I the pool is not ideal for lap-swimming.