The guy who was walking me through tree maintenance in the yard mentioned it offhand and I filed it away in the back of my mind. A few months later, after some reading about it, I bought a tree tapping kit off Amazon that seemed easy enough to implement and kept a watch on the weather forecast to find the right day to tap the trees.

The kit had three taps and buckets, so I tapped three trees. Tree #1 is my superstar, producing the most sap and with the most sunlight during the day with the most space. Trees #2 and #3, who are known as the twins, are partially shaded and get less sun during the day. It was three days from the time I tapped it that all three buckets were pretty much full. Probably about seven gallons of sap.

I grossly underestimated how long it would take to boil that amount of sap! I was very confident with Meg that it would take a few hours on a Sunday and we’d have plenty of time to relax with dinner before I sat down to watch the All Elite Wrestling pay-per-view Revolution. I started around 2 pm and it boiled. And boiled. And boiled. And boiled. I was using a turkey frying kit that I picked up at Home Depot, which has a pretty small surface area, meaning it would take longer to boil.

Thankfully, it was a warm night and I was able to hang out outside, enjoy the night, the pay-per-view, and keep resetting the turkey fryer’s automatic safety timer every fifteen minutes. I think the boil was mostly done around 11 pm. I jarred what was left and would handle it in the morning.

In the morning I boiled down the rest from about 4 cups to about 2 cups. It filled the house with a great smell that soon became overwhelming and required all the windows to be opened. About an hour later, I had syrup!

Ultimately, it’s syrup! It’s amazing that I can just make this stuff myself! I haven’t used it yet but I spent the day walking to the fridge and dipping a finger in for a taste of it. I named it after Adam Page, the current AEW heavyweight champion, since I watched him retain his championship while it was finishing the larger boil.

I started the second boil the next day (Tuesday) but after a disaster. It was especially windy and Tree #1’s bucket was nearly full and the weight combined with the wind pulled it right off the tree. The twins produced enough to get a small batch going. I still had daylight, so I worked for about two hours outside while the boil got to a good enough place to bring it inside to finish up. A bit too chilly to really enjoy the day but I brought it inside before dinner.

I filtered the sap multiple times, probably once more than was needed but I wanted to see if I could produce something more clear than the first batch. In honor of my sap disaster, I named the second batch Dr. Bunsen Honeydew.

I still have some work to do and there are some tools that can help me get this done a bit easier than it has been, but it’s still been fun. The amount left on the second boil and the lighter color makes me think that I probably didn’t boil it enough but the taste seemed about right. To really test it, I’d need a refractometer to see the sugar content. Not sure if this batch needs another filter or not but I’m happy with this round too. I’ll need some smaller jars too if I’ll keep doing this every few days.

The black line shows Lombardi’s Yelp score over time, the green horizontal line shows the average score as of September 2019. In September of 2006, Lombardi’s hit 4 stars and hasn’t changed since. You can see from the black line that the average has been slowly trending towards 3.5 stars, but the rating was 0.0385 stars above the threshold. The rolling mean tells a more complete story, having drifted into the 3.5 star territory in 2015.

How many of those 6,000 reviews are quality reviews?

In order to see the impact of filtering reviewers, I took a look at the Yelp reviews of Young Ethel’s, a bar that I love in Brooklyn, that has 19 reviews and 5 stars. 15 five star reviews, 4 four star reviews, a 4.7895 star score.

Since Young Ethel’s is relatively new, we can assume that their friends and family will write some reviews to help them get started.

A few red flags for this review. No friends, no user photo, a low number of reviews. I know the opening date of the bar (August 22, 2019), so this is also within a month of opening.

Of the 19 reviewers, 3 are missing user photos and 3 have no friends. 5 have neither.

All reviews: 4.7895 stars

Reviewers with Photos: 4.75 stars (-0.0395 difference)

Reviewers with Friends: 4.75 stars (-0.0395 difference)

Reviewers with Photos and Friends: 4.7143 stars (-0.0752 difference)

Filtering out the reviewers that have no photos or friends, we’ve lowered the star rating by half a star. There’s one more group I believe that should be filtered out, which are Yelp Elite Squad members. Elite members are rewarded for interactive on the site with exclusive events. New businesses would be prime for new reviews from Elite members. Since I’m looking to filter out reviews that aren’t genuine, I’d be remiss in leaving Elite reviews in the score.

Non-Elite Reviewers with Photos and Friends: 4.8 stars (0.0105 difference)

By removing Elite reviews, the score is up in comparison to the score. Without the Elite members and suspect reviewers, there are 10 reviews left.

My next step would be to see what impact this has on a larger scale, but for now, it’s interesting to see how weeding out reviews can impact the overall score.

not

list = [None] * len(old_list)


Use Case:

When I create DataFrames, I’ll create a series of lists to fill. The DataFrame is created by using lists. reviews is a list of HTML blocks. Here’s how I would have normally set up the lists.

texts = [None] * len(reviews)
dates = [None] * len(reviews)
ratings = [None] * len(reviews)
checkin = [None] * len(reviews)
photos = [None] * len(reviews)

A quick timing shows that texts = [None] * len(reviews) is about 22% slower than using a defined variable as the length of the list.

Time to create five empty lists with a length of 20:

• Using review_len: 772 ns ± 5.66 ns
• Using len(reviews): 969 ns ± 13.5 ns
• Using 20: 706 ns ± 14 ns

I’m surprised to see that using a definitive length, 20, didn’t save more time.

Saving nanoseconds may seem like a trivial amount of time, but continually attempting to optimize code is admirable.

Use Case:

I have a DataFrame with two sets of dates and the difference between them. There are two different ways to get a rounded up number of days, depending if you want a Timedelta or an integer.

Method 1: Integer

>> df.delta.mean()
Timedelta('21 days 15:00:00')
>> df_ex.delta.mean() / pd.Timedelta(days = 1)
21.625

That’s the answer, but also for my purposes, partial days is too fine for my purposes. Dividing a Timedelta by pd.Timedelta(days = 1) will result in an integer.

>> math.ceil(df_ex.delta.mean() / pd.Timedelta(days = 1))
22

Method 2: Timedelta

>> df.delta.mean()
Timedelta('21 days 15:00:00')

>> pd.Timedelta.ceil(df.delta.mean(), freq = 'd')
Timedelta('22 days 00:00:00')