Why (Project) BASS Matters

Published October 12, 2020, 1:53 PM

by Wilson Chua

“Hawthorne effect: You measure what you manage”.

“I don’t think you knew at that time when DICT had a MOA with BASS, was to use BASS to check compliance of the third major Telco player on its commitments, in a crowd sourcing sort of monitoring.

As you know the selection process was thru a Highest Committed Level of Service (HCLOS), the first time to be done in the world. HCLOS can only be effective if independently monitored.

In fact, even the incumbent Telcos are also improving their services because of BASS monitoring. I hope DICT will continue to take advantage of BASS for this purpose,”- ex-DICT head Eliseo Rio Jr

How it started:

ProjectBASS was born out of an FOI.GOV.PH rejection letter. I had requested data from the National Telecommunications Commissions covering cellsites and their spectrum usage. The NTC denied the request.

Since I wanted to find out how efficient the Telcos were in using their assigned frequencies, I sought other ways to get this. Maybe crunching mobile usage data might reveal some insights that can improve Philippine internet?

Never one to give up, I spent the next weeks thinking about this. One fine day — while chatting with Benjamin Tan (who sadly passed away) we hit upon a novel idea. The smartphones we are using have GPS and other tech gizmos built in. Can we convert these phones into mobile internet speed testers? It would be cheap as we only need to install a mobile app. It would save millions for government regulators. It should be possible (theoretically) to crowd source the speed tests.

My apprehension then was we needed a lot of things to fall in place for this to work. We needed a hot shot core developer team that can create the mobile app. We also needed a social media team to recruit supporters/volunteers. The volunteers in turn provide the bulk of the effort to help measure Philippine internet.

“Alone we can do so little. Together we can do a lot”

The core team volts in:

I tested the waters by writing an article calling for kindred spirits in Manila Bulletin. Who, I asked, would be keen to help out? And literally within a few hours, Paul reached out via Facebook Messenger.

Paul Sydney Orozco had already finished coding a speed testing mobile app. It was already posted on Github. He donated his code to ProjectBASS. Melby who was an intern at the time, went to help Paul tidy up the code and we had a mobile app in time for the Philippine Internet anniversary!

Ms Fleire Castro (queen of social media of the south) and Ms Olay Rullan also joined in. I greatly appreciated their social network efforts to spread the word. And promote it they did with gusto. Ms Fleire’s team also crafted the website, logo and core presentation for ProjectBASS.org. Olay’s network of friends helped with awareness and fielding technical project questions.

Along the way, Melby introduced us to A-Ar Andrew Concepcion. I always wondered about that “A” part. Was he named like this to get his name to be always listed on top alphabetically? With that awesome trio of Paul, Melby and A-Ar, we soon had added features to the mobile app. Then Olay recruited Andrew Alegre to help ProjectBASS roll out the IoS version of the app.

Unexpected supporters come in :

Manila Bulletin Editor Art Samaniego also pitched in by organized a coming out event with the media outlets and telcos. I remembered that among the Telcos, only Globe representatives showed up. No one from SMART came.

In the months that followed, Mr Tzar Umang joined our ragtag group and helped with his vast experience working in Science and Tech and Government. Truly an A team was assembled to this day!

DOST-ASTI came in and provided support in the form of server and bandwidth from PHOpenIX. As the measurement point, PHOpenIX is a central and neutral location. This makes it a fair basis for measuring major ISPs’ bandwidth performance.

One memorable support came from Yugatech. When the site posted about ProjectBASS, we got so much measures that our server/s almost died. In fact I think one went to heaven for a few hours and came back.

Then of course, there was Prof Rom Feria of UP Diliman who with his entire class volunteered to do a data privacy code check. They found that we were collecting IMEI (International Mobile Equipment Identity). Hackers could theoretically use this IMEI to clone your phone.

What can we do? Fortunately, Dondi Mapa a friend and husband of my UP diliman classmate Cherry Gonzales Mapa stepped up and helped. Dondi was with the NPC at the time. He suggested that we drop the last half of the IMEI. This allowed us to retain the important information (first half of the IMEI) without revealing personally identifiable data. What a brilliant and elegant solution. Thanks Dondi!

Winthrop Yu, Grace Mirandilla Santos and Achie Atienza (ISOC PH, Better Broadband and PHNOG) collectively put the prestige of their orgs behind ProjectBASS. Thanks guys!

We caught the attention of then Acting Sec Eliseo Rio Jr at DICT. He really moved a lot of bureaucracy to have ProjectBASS.org dashboard embedded into the dict.gov.ph website. (This agreement has since expired). Little did we know that the Secretary leveraged ProjectBASS for their first of a kind HCLOS basis for determining the 3rd Telco. DITO won the bidding.

The benefits of data collected:

ProjectBASS collected bandwidth, cellsite tower ID, signal strength together with our volunteers smartphone make and model. If they connected via WIFI, we also captured the WIFI SSID. This later on would prove useful for tracking the PH Government’s Free Wifi program. Who would have thought about this?

The data our volunteers collected for us is now being used to help the Philippines. Some in expected ways. Others in totally UNEXPECTED ways.

a. Improving Internet access
To date, 3 ISPs discovered line errors that degrade internet service because of the data visualizations and BASS WIFI rankings.

Top WIFI Networks arranged by Region, Province and Municipality

The 3 ISPs started out by calling our attention to a discrepancy. Why, they asked, was ProjectBASS measures LOWER than the OTHER speedtest they were using? So we sat down and did some analysis. Turns out there were packet losses on the lines. For some unknown reasons, their OTHER speedtest always report suspiciously high bandwidth despite these packet losses.

Going back to the story. I like interacting with technical people. They see the truth when they see the proof. Once they discovered the packet losses, it was an easy matter to fix the root causes and voila! These 3 ISPs ProjectBASS scores went up.

That benefited not only the ISP operators, but also their consumers/subscribers. And increasingly, these subscribers are consulting the Top WIFI dashboard (featured above) to help them make informed decisions as to which service to sign up for. This kicked off a virtuous cycle. As more subscribers base their subscription info on data, the more Telcos are incentivized to be better than their competitors.

ProjectBASS presentation for Singapore’s 4th Telco (MNVC) Circles Life

b. Driving Online Education

When Covid hit and school children took to online lessons, one key question in all our minds was “Do we have enough bandwidth?”

“ProjectBASS data to the rescue!”
We already have the cellsite location and median bandwidth. I thank PHILGIS for their barangay shape files. Together with DEPED school location data, we can provide stunning visualizations. These visualizations and data can help make data based decisions.

Higher Density Cell Sites in Metro Manila with schools in Red House Icons
Lower Density Cellsite somewhere in Mindanao

Notice how in Metro Manila, each school has access to several cell sites? While in Mindanao, it is the other way around. Several schools have access to only ONE cell site. This is what the DICT is working hard and fast to address.

c. Paving the way for (Tele)-Health

Can we do Telemedicine in times of covid?
Thanks to Martin Gomez and his wards (Sean Chua, Adrian Serapio, and Conner Manuel) for cleaning up the data and geo-coding each of the Health facilities’ GPS data. We can use this to visualize the answer for the community.

Each of the health facilities are marked by GREEN Cross on the map. And the QGIS algorithm showed the nearest cell site that they can connect to. Hmmm….Not good.

Not only are health centers so far away from the nearest cell site, they have to also share the meager bandwidth from the closest cell site.

Somewhere in Zamboanga, Philippines

d. Data Driven Economy (World Bank Report)

I am still feeling pretty stoked about this. No less than the World Bank’s Philippine Digital Economy 2020 report picked up and cited ProjectBASS data for this chart:

How YOU can help:

Looking toward the future, we need to sustain and keep on improving our service to our community. Here are some of our plans:

a. International IX hosting

Can kindred souls out there connect ProjectBASS to International IX (internet exchanges) ? We sure could use free hosting to add more measurement locations. These measurement locations enable ProjectBASS to provide better insights on a per continent or per country basis. You may want to be the country lead too.

b. User/Volunteers specially in far flung areas

We never have enough volunteers…specially in far flung areas. The more remote you are, the more our community needs YOU!! Just go over to Projectbass.org and install the mobile app to your phone then start measuring the internet at your location.

c. Technical gurus to develop new features

As the times change, we identify new features that can help our community. I have leaned on and extracted so much out of our core developer team that I hesitate to put more on their plate. If you have the skills and willing to contribute to improving Philippine Internet, we need your help with the following :

c1. DeadSpot reporting

How do you measure internet when you don’t have internet coverage? If you have a nifty way to help spot these dead spots, we need you!

c2. Frequency data collection

Can smartphones also collect the frequency and tie it back to the cell site? We can use this data to analyze the frequency utilization. This insight can help Telco planners and government regulators (DICT) to develop policies that drive better Internet.

c3. API for embedding

We are mobile. An API (applications programming Interface) for digital businesses can be a win win proposition. They can embed ProjectBASS into their app. Their app users are turned into ProjectBASS volunteers. And the app owners get free bandwidth speedtests!

There are a lot more to say, but if you have reached this point in the article, I know you are keen to help. Drop us a line. Visit our FB page and LIKE it. https://www.facebook.com/ProjectBassApp