2026 - Against Sunnyvale Measure G

This is a place holder until after the voter's information guide is finished.

Argument against Sunnyvale''s Proposed New Library

 Does Sunnyvale Need a New Main Library?

Sunnyvale’s lovely main library has been serving the community since 1961, through over 60 years of family use, weather and earthquakes.

Summary

Sunnyvale (California’s 37th largest city) proposes a $290,000,000 bond (Measure E) to tear down the existing library and build a 120,000 sq. ft. new one, double the 60,000 sq. ft. of the current main library, making it the 7th largest library in the state. The 25 year bond at 4% will add $169M so total cost will be $459M.

(Ref: https://ca.countingopinions.com/pireports/report.php?e1af9d8fefa2ef3389ff88b2c9cdd028&live )

For an amazingly good AI-generated podcast summarizing this blog done by Google's NotebookLM, click on this link:

https://notebooklm.google.com/notebook/19a06bfb-24f5-48bc-b9cf-72744e062687/audio )

I love libraries. After I was elected to Sunnyvale's city council I was happy to vote for the new 20,000 sq. ft. branch library in north Sunnyvale which just recently broke ground (no new taxes required). I would support a new branch library, especially one for children like Palo Alto's (link opens in new tab).

Interior of Palo Alto Children's Library
https://library.cityofpaloalto.org/locations/C/

But for the following reasons I must oppose the new library bond proposal - Measure E.

Library usage and circulation in Sunnyvale, Santa Clara, and around the country is down a lot as you can see in the charts below - 50% drop in patronage nationwide! LA down 74%, San Francisco down 65%, etc. Sunnyvale down 62%! Go to Sunnyvale's library even on weekends, and it has lots of empty desks and chairs. When the new Sunnyvale branch library opens, the main Sunnyvale library will see even lower patronage. So why build a new one?

US library patronage (blue lines) is down 50%.  Physical circulation is down almost as much.  Yet eBooks (gray lines) are soaring. (click on image to expand)
Source: https://www.publishersweekly.com/pw/by-topic/industry-news/libraries/article/95383-the-quiet-crisis-facing-u-s-public-libraries.html

Sunnyvale peak usage was about 900,000 in 2008/2009.  Current usage is 38% of that peak at 337,831

The proposed cost of over $2400/sq. ft. is well over double the cost of new libraries being built around California. That includes Sunnyvale's own new branch library which just recently broke ground at barely above $1000/sq. ft. New libraries in CA now (2024) are generally around $1000 to $1200 per sq. ft. (see "Cost Comparison" below). Even at $1200/sq. ft., adding on 60,000 sq. ft. by an addition would cost "only" $72M, saving $218M - we could do a lot with that for public safety, parks, schools, and helping those in need. The most expensive place for construction of offices (and libraries) is $1270/sf in NYC. (see https://www.clarisdesignbuild.com/2024-update-commercial-construction-cost-per-square-foot-in-the-us/). 

As much as I love libraries, bigger is not necessarily better. All data on usage shows that easy access is absolutely the most important factor. For example, Fresno (population 520,000) has an 82,700 sq. ft. main library yet a branch library of only 22,000 sq. ft. has nearly double the circulation at only about 1/4th the size.

Fresno's Woodside Branch Library 22,000 Sq. Ft.
But nearly 2X the circulation of the 83,000 sf main library
(click image to enlarge)

Even smaller, at only 10,000 sf, Fresno's Fig Garden Branch had a circulation over 20% greater than the 8X larger main library. It is a rented commercial space in a strip mall - small, but kids and parents with strollers can walk to it, and it has easy parking for families. Access is key. More on these libraries here: David and Goliath (link opens in another tab).

To increase library usage, especially by kids and families, more branch libraries they can walk to is the answer, not gigantic monuments.

At the time of the first library bond proposal in 2007 (more at "History", below), local libraries were absolutely packed, even on week nights. It is dramatically different now. The change at the Sunnyvale and Santa Clara libraries is obvious. In room after room with seating for 28 or more, there are only 4 to 6 people. A desk that seats six has only one person, if that. Even on weekends, there is no shortage of seating - or parking.

If Sunnyvale wants to enhance library usage, the best way would be branch libraries in underserved areas (see "Alternatives", below), possibly sharing the cost and space of the nice libraries in the middle schools and high schools. Some cities do this with city librarians taking over after school ends and on weekends. A win-win for cities and schools!

Campaign Contributions

It is no accident that the biggest contributors to the "Committee to Support City of Sunnyvale Library Bond Measure" are architects in San Jose and San Francisco, and contractors to the city. In a weird little operation, the N. CA Carpenters PAC recently gave $5,500 to Charlsie Chang's campaign. She has no opposition so why does she need contributions? Anyway, as of this writing she has so far given $2,000 to the Support Sunnyvale Library Bond Measure campaign. C.f., City of Sunnyvale Electronic Filing System (netfile.com)

Cost Comparison

$290M for 120,000 sq. ft. = $2,417/square foot. That is well over twice the cost of Sunnyvale's new branch library in the Lakewood district (still under construction) which cost $23,000,000 for a 22,000 square foot library - just over $1,000/sq. ft. as reported in the San Jose Mercury News: https://www.mercurynews.com/2024/06/07/new-library-comes-to-north-sunnyvale-replaces-old-community-pool/

In 2022, Santa Cruz priced a new 38,000 sq. ft. library at $1,118/sq. ft. as reported here: https://lookout.co/measure-o-new-library-vs-renovated-library-what-would-santa-cruz-get/

Riverside is building a 15,000 sf library for $20M = $1,333/sf as seen here: https://riversideca.gov/library/about-library/new-spc-jesus-s-duran-eastside-library-building-project

In Contra Costa County Bay Point's new library of 21,000 sf will cost $20M = $952/sf as reported here: https://contracostaherald.com/contra-costa-county-library-awarded-9-9m-grant-for-new-bay-point-library/

A library is similar to an office, and in 2024, office space construction costs are highest in New York City at $1,270 per square foot

(see https://www.clarisdesignbuild.com/2024-update-commercial-construction-cost-per-square-foot-in-the-us/). 

The big beneficiaries of this $290M bond will be big construction companies.

How Big Should a Library Be?

It's nice to know the size of libraries near Sunnyvale to get an idea of how big the proposed 120,000 sq. ft. library would be. Sunnyvale's main library is 60,800 sq. ft.

The proposed new library would be twice the size of the main libraries of Mountain View, Cupertino, Milpitas, and would be 50% bigger than Santa Clara's.

Here are the sizes of nearby libraries, all of which by state law give library cards to anyone - without charge.

Santa Clara: 80,000 sq. ft.

Mountain View: 60,000 sq. ft.

Cupertino: 59,600 sq. ft.

Milpitas: 60,000 sq. ft.

Fremont City: 68,500 sq. ft.

You can see most libraries find around 60,000 sq. ft. convenient.

There's a reason for that. What is most effective in making books and other materials available to people, especially kids, is easy access with branch libraries as discussed above.

If one library doesn't have the book or e-book you want, you can get it at another city library nearby, or the local one will get it for you from the county.

(Source: CA state records: https://ca.countingopinions.com/pireports/report.php?e1af9d8fefa2ef3389ff88b2c9cdd028&live )

History

Back in 2007, Sunnyvale proposed replacing the main library with a brand new one, at $1,000/sf. Yet again, this was twice the cost of the average of other new libraries around the state at the time, including in the SF Bay Area. We showed then many examples of lovely new libraries built in CA around the same time for under $500/sf - half Sunnyvale's 2007 proposal of $1,000/sf. In the linked articles we show the lovely interior of Camarillo's Library (at less than half the cost of Sunnyvale's 2007 proposal) (LINK).

Exterior of Camarillo's Library
(click on photo to enlarge)

as well as the libraries of Alameda, Calabasas, etc. and others here.

Exterior of Calabasas' Library
(click on photo to enlarge)

Alternatives:

We could add a branch library like the new 20,000 sq. ft. library Sunnyvale is building in the Lakeview district. Many people in other areas are closer to nearby libraries like Cupertino's or Santa Clara's than to Sunnyvale's main library but a branch library near Sunnyvale Middle School or Peterson Middle School might be welcome.

Or we could modify the current library. Perhaps add a 2-story wing of 20,000 - 30,000 sq. ft. for a total of 80,000 (like Santa Clara's) or 90,000 (50% larger than the current library).

Sunnyvale's Public Library

from above

Sunnyvale City's website makes arguments about square feet of library space per resident, but that is a false argument because every library has a minimum requirement of entry way, help desk, hallways, etc. A city twice as large as another doesn't need twice as many entry ways, help desks, etc.

Even if we did want to increase the size of the main library, wouldn't it make more sense to simply add on to the existing library?

Destroying a perfectly good building instead of adding on is an environmentally destructive waste of money and material.  We recycle food waste and bottles but trash perfectly good buildings?  Adding rooms, amenities, solar panels and insulation to the existing building as we do to our homes is far better than trashing a fine old building.  The saved money could be far better spent helping the homeless, providing better roads, and public safety.

Sunnyvale residents can and do check out books, eBooks and media from all community libraries near home and work at no charge.  Is it worth $290,000,000 just to say “my library is bigger than your library” ?

Sunnyvale’s city council needs to pay less attention to what builders and construction unions want and more to what Sunnyvale really needs.

St. Gallen Abbey Library, Switzerland
Maybe cheaper than the over quarter $Billion Measure E?

Manhattan, NY Commute Flows

Manhattan, NYC

New York City is the city Americans think about when they compare city life with suburban or rural life.

And, when people advocate for high-density living, public transit, and against suburban "sprawl" they usually use NYC as their model, for better or worse.

Permalink to this post for sharing:

https://meetingthetwain.blogspot.com/2021/08/manhattan-ny-commute-flows.html

Manhattan Subway

And, when they think of NYC, they are usually thinking of Manhattan - one of the five boroughs that make up NYC.  The borough of Manhattan (New York County) is 24 square (land) miles of tall buildings that is the original New York City (the other four boroughs were added over time until 1898).

Those advocating for density often cite NYC (Manhattan really) as the model to be emulated.  Much of what they think about NYC isn't true, particularly commuting.  Lets look at the numbers.

Commute Flows

Summary:

Manhattan, with only 24 square miles of NYC's 305 sq. miles, is the job center for the NYC metro area.  With over 2.3 million jobs and only 734,000 resident workers, Manhattan has more than 4.5 jobs for every resident worker.  Nearly 80% (77.8%) of those who work in Manhattan commute in from other areas.

One would think that nearly all residents could find a job they like within an easy commute by walking, bus, or biking.    But, in fact, nearly 30% (29.7%) of those who live in Manhattan (one of the most expensive places in North America) commute out of it to work in places which are almost always cheaper to live in than Manhattan.

Looking at the changes over the 16 years from 2002 to 2018, we see an increase in commuters.  There is an increase in both the number and percentage of those who live in Manhattan commuting out.  Not only that, they are commuting from Manhattan further and further out.  It is a slow, steady increase.

Nonetheless, as of 2018, only 15% of Manhattan residents commute more than 10 miles from home, although that is slowly but steadily increasing.

Commute Flows:

From the US Census Tool "OnTheMap" we have the commute flows of Manhattan in the first image below.  This shows that 1.8M workers commuted into Manhattan, while 217,911 living in Manhattan commuted out, and 515,641 of those living in Manhattan worked there.

Figure 1 - Commute Flows 2018

(click to enlarge)

In Figure 2, below, we have the commute flows of Manhattan in 2002, 16 years earlier.  This shows that 1.4M workers commuted into Manhattan, while 146,013 living in Manhattan commuted out, and 436,580 of those living in Manhattan worked there.

Figure 2 - Commute Flows 2002

(click to enlarge)

Let us compare this with 2002 - 16 years earlier - and see what changed over time.

Below is the data from 2002.  Key points are that during those 16 years from 2002 to 2018, the number of jobs in Manhattan has increased by 512,404 (28.3%) while the number of those living in Manhattan increased by 150,959.  The number living in Manhattan and also working there increased only by 79,061 (18.1%).  Over the same period, the number of those living in Manhattan and commuting out increased by 71,898 (49.2%).  In other words nearly half of the increase in population of workers living in Manhattan during this period commuted out.  This resulted in an increase in the percentage of those living in Manhattan and commuting out from 25% to 30%.

Paradoxically, the addition of workers living in the job center resulted in a larger percentage commuting out of Manhattan - from 25% in 2002 to 30% in 2018.  We see this in figure 3 below:

Figure 3.  Commute flows - 2018 vs 2002

(click to enlarge)

Commute Distance and Direction:

In Figure 4 we see the changes from 2002 to 2018 in where and how far workers residing in Manhattan commute.  In that 16 year period, those commuting more than 10 miles to work increased by 40,000 - from 77,000 (12.1%of resident workers) to 107,000 (14.5% of residents), an increase of 39%.

Figure 4: Manhattan Residents

Commuting Distance & Direction

(click to enlarge)

In Figure 5, below, we see the changes from 2002 to 2018 in how far workers commuting into Manhattan travel.  In that 16 year period, those commuting more than 10 miles to work increased by 162,000 - from 698,000 (39%of non-resident workers) to 860,000 (37% of non-residents).  The percentage decreased but the sheer number increased.

Figure 5: Manhattan Non-Residents

Commuting INTO Manhattan - Distance & Direction

(click to enlarge)

The places that people commuting out of Manhattan is also available from "OnTheMap" by distance range.  The 10-24 mile commute range is shown below.

Figure 6: 10-24 Mile Commutes of Manhattan Residents

Shown in Purple

(click to enlarge)

Remarkably, workers are commuting from the most expensive major county in the US to much cheaper suburban areas, usually considered "bedroom communities".

Congestion and Time Lost to Commuting

The result of this density and concentration of jobs and housing has been some of the worst congestion in the world.

Those arguing that high density results in less commuting need to deal with the reality that the densest county in the US has seen a significant increase both in the number of commuters, and in their miles traveled to work.

The reality is that nearly 4 out of 5 of those who work in Manhattan commute from outside - often the suburbs or one of the more suburban parts of NYC such as Staten Island.  New York commutes take the longest time in the US.

Figure 7 - Commute Time

(click to enlarge)

Fortunately, NYC has an excellent (and expensive) public transit system.  Still congestion is worse than almost all other cities in the entire world.  The following chart from the Economist shows that in time lost to congestion, NYC is second only to LA, and tied with Moscow.  In terms of cost of congestion per driver, NYC is the absolute worst in the world.

Figure 8 - Commute Time Lost to Congestion

From the "Economist" - 2018.
https://www.economist.com/graphic-detail/2018/02/28/the-hidden-cost-of-congestion

Conclusion:

Increased density does not result in better commuting.  It simply concentrates the congestion, and distributes the concomitant pollution to the suburbs.

Ridiculous RHNA ("Ree-na") - Part 3

Ridiculous RHNA ("Ree-na")

Part 3

All kinds of housing for all kinds of people

(Link to this post for sharing:

https://meetingthetwain.blogspot.com/2021/04/californias-2021-rhna-bills-part-1.html)

Summary:

We show that only 5% of CA cities and counties build as many houses as projected by their recent "Regional Housing Needs Allocation" goals. The other 95% didn't "make their RHNA numbers" and are therefore subject to "streamlining".  ("Regional Housing Needs Allocation" is abbreviated RHNA and pronounced "Ree-nah".)

Examples of jurisdictions large, medium and small show CA's Housing and Community Development agency's (HCD) RHNA numbers imply dramatically unrealistic growth in housing needs which therefore cannot be met by cities and counties.

Recent public comments in favor of raising RHNA numbers suggest that intent of high RHNA numbers is not to add more housing but to remove any restrictions towns and counties might put on building.

Other posts related to RHNA:

Part 1: https://meetingthetwain.blogspot.com/2018/09/ree-diculous-ree-na-part-i.html, 
Part 2: http://meetingthetwain.blogspot.com/2018/09/ree-diculous-ree-na-part-ii.html

Background on RHNA

RHNA was established decades ago.  The idea was that California's increasing population needed more housing.   To assist cities and counties in planning for that housing the California Dept. of Finance estimates population growth which then become the basis for projected housing growth (RHNA numbers) for a region.  The RHNA numbers are administered by California's "Dept. of Housing and Community Development" (HCD).

An excellent history of the perversion of RHNA for political purposes is here:

https://www.berkeleyside.org/2021/10/25/berkeley-housing-rhna-development

Cities & counties don't actually build housing but simply authorize others to build it.  The only thing cities can do is zone land to accommodate projected growth.  Once the land is appropriately zoned to accommodate projected growth there is nothing else the city can do.  

Until recently, if no one built housing according to the RHNA numbers, HCD didn't do anything.  If the projected growth didn't happen, it was not presumed to be the fault of the city. That has changed with new legislation which implicitly assumes that cities can somehow induce buildings to be constructed by private development companies, or that it is only the cities that are holding back more housing from being built.

We show here:

1. that RHNA numbers assume a rate of growth far higher than that of the last 30 years,
2. that housing construction has kept up with population increase,
3. that there are more housing units than households in every county in California,

We also propose ways to modify RHNA numbers to keep it line with realistic growth.

These facts are important because the proposed bills in the California legislature are based on the assumption that RHNA is a true assessment of future growth.  Therefore, the reasoning goes, if RHNA numbers are not met by a jurisdiction it must be the fault of the the jurisdiction.

Taking these items in sequence: 

                               RHNA Numbers are Unrealistic

California has 58 counties and 482 cities and towns (there is no legal distinction between a city and town) for a total of 540 jurisdictions.  Of those 540 cities and counties, HCD's 2019 "SB-35 Statewide Determination Summary" lists only 28 jurisdictions that comply with the RHNA numbers assigned them.  So, out of 540 jurisdictions only 5% (one in twenty) met their RHNA numbers.

Image 1: Intro segment of HCD 2019 SB-35

HCD's SB-35 Statewide Determination found at:

https://www.hcd.ca.gov/community-development/housing-element/docs/sb35_statewidedeterminationsummary.pdf

A look at the entire list of 28 jurisdictions shows that those meeting their RHNA numbers are small cities representing a tiny fraction of California's population.  RHNA numbers for counties do not refer to the entire population of the county but only the unincorporated parts of the county.  In populous counties like Santa Clara (nearly 2 million people) almost all of the population lives in incorporated cities which have their own RHNA numbers - the population under county control is very small.

-------------- +++ -----------

Some of the rationale given for assigning RHNA numbers is 'to ensure that all jurisdictions contribute to alleviating the housing shortage'.  The RHNA numbers for cities and counties are unrealistic and shows no relation to either historic growth patterns or any housing "shortage".  We show examples of small, medium and large jurisdictions for which RHNA is unrealistic.

Examples of extraordinarily unrealistic RHNA allocations

At the small end, consider Alpine County.  It borders Nevada just South of Lake Tahoe.  It had a total population of 1,200 in 2000, 1,175 in 2010 (decrease of 25) and an estimated 1,129 (decrease of 46) residents in 2019.  Over 96% of the land in the county is owned by the US as a National Park or National Forest.

Image 2:  Alpine County: Population 1,129 & Declining

5 housing units for every household

Failed to meet their RHNA goals and therefore subject to monthly fines up to $100,000.

According to the US Census, Alpine County (2019) comprised 350 households.  Home ownership is 84% as of 2019 compared to 64% for the US.  There are 1,782 housing units in Alpine County, i.e., about 5 housing units per household - meaning there are 1,432 are mostly empty vacation homes.  I.e., there are 4 empty housing units for every occupied one.

Image 3:  Alpine County: RHNA 

Alpine County RHNA mandated 11 "Above Moderate" housing units.

If only 10 were built Alpine could be subject to monthly fines up to $100,000 under AB 215.

With a population declining over 19 years, far away from the expensive coastal cities, almost everyone owning their home, and five housing units per household what is the possible justification for assigning any RHNA requirements at all to Alpine County?

Image 5:  HCD Determination

Not enough "Above Moderate Income" housing built in

298 Cities and counties so state laws for those that didn't meet their RHNA numbers kick in.

From: https://www.hcd.ca.gov/community-development/housing-element/docs/sb35_statewidedeterminationsummary.pdf

On Page 2 of 6

For Alpine County it is clear that RHNA devoid of any rational purpose.  Are there any other counties in a similar situation?  It turns out that there are a baker's dozen counties like Alpine County with small and declining populations (from 2010 to 2018), excess housing units, and high home ownership.  Throw in two more that have trivial increases and we have 15 rural counties without a housing problem yet potentially subject to high fines under Chiu's AB-215.

13 Counties Losing Population (plus 2 that didn't gain much)

1. Sierra County              (33)  Population decreased by 33
2. Alpine County             (21)  Population decreased by 21
3. Calaveras County      (421)         "          "                  421, etc. 
4. Del Norte County    (1,389)
5. Kings County          (1,320)
6. Lassen County        (3,984)
7. Modoc County             (74)
8. Mono County             (380)
9. Mariposa County       (122)
10. Plumas County          (234)
11. Siskiyou County         (288)
12. Trinity County            (151)
13. Tuolumne County      (625)
14. Amador County             +3    Population increased by   3
15. Inyo County                 +31    Population increased by 31

AB 215 would allow the imposition of monthly fines of from $10,000 to $100,000 on these counties.  For Alpine County, $100,000/month works out to about $100 per month for every man, woman and child or about $3,600 per year for the typical household - probably more than their property tax - solely because no builder decided to put up housing in a sparsely populated county with a 5-to-1 housing surplus, and a declining population where 85% own their own home.

-------------- +++ -----------

For small jurisdictions, consider Amador City.

There are many small towns in California with declining populations as well.  Consider the town of Amador.  US Census 2019 population estimate of 190 down from 195 in 2000.

Image 4:  Amador City: Population 190 & Decreasing

Amador is among the 95% of California cities and counties that failed to meet their RHNA.  As in Alpine County builders didn't construct enough "Above Moderate Income" housing units so they get the full brunt of state law, just like Alpine County - see image 5 below from 2019 HCD.

Image 5:  HCD Determination

Not enough "Above Moderate Income" housing built in

298 Cities and counties so state laws for those that didn't meet their RHNA numbers kick in.

-------------- +++ -----------

For middle-sized jurisdictions, consider Pasadena.  

Image 6:  Pasadena, California

Home to the Rose Bowl

Since 1990 it has grown from 131,591 residents to 141,029,  i.e., an increase of 9,438 residents over 30 years or 315 residents per year.  It now faces a RHNA of 9,409 housing units over the next eight years.  This means HCD's numbers imply housing for an additional 23,522 residents (based on Pasadena's current household size of 2.5 people) or 2.5 times more residents in the next 8 years than they have increased in the last 30 years.  In 8 years that averages 2,940 more residents every year from RHNA vs. the actual 315 annual population increase over the last 30 years.  This is nearly 10 times more growth than the historical average!

-------------- +++ -----------

For large jurisdictions, consider Los Angeles, California's largest city.  Between 1990 and 2019 it grew from 3,485,398 residents to 3,979,576,  i.e., just over 500,000 residents in 30 years or about 16,700 new residents per year.  LA's RHNA allocation for the next 8 years is 82,000 housing units as seen in LA's City Housing Element Plan (page 79).

At LA's current 2.8 person household size this implies a growth of 229,600 residents in 8 years or 28,700 residents per year vs. the 16,700 residents per year over the last 30 years.  This is an addition of 12,000 more residents per year than the 30-year historical average.  In other words, RHNA projects growth 170% above the 30-year actual growth rate.

Image 7:  Los Angeles, California

Nearly four million residents

Los Angeles

Growth of 16,700 residents per year over the last 30 years.

RHNA implies 28,700 residents per year over the next 8 years.

Los Angeles in 2015 had 1,478,666 housing units for 1,383,869 households for a surplus of  95,000 housing units, i.e., nearly 7% more housing units than households.

Reason for Unrealistic RHNA numbers

The following SF Chronicle article from July-August 2020 makes it very clear that the point of higher RHNAs is to make sure cities and counties can't meet them.  

Bay Area needs to build lots of housing to meet state goals - and goals called too low

Bay Area needs to build lots of housing...

Quotes from article:

“If we got a RHNA number in line with the Southern California number, we would increase another 50%, which would mean more counties would not be hitting their number, which would mean that housing would be streamlined in those communities,”

"Matt Regan, who heads up policy at the Bay Area Council, a pro-business trade group, said he supports the highest possible number so that developers will be able to get quick approvals by doing 10% affordable."

When jurisdictions fail to meet their unrealistic RHNA goals, state mandated "streamlining" kicks in.  This means no public notices are required, and builders can put in only 10% "affordable housing" instead of city mandated numbers which are typically 15% or higher.  There are also reductions or eliminations of off-street parking requirements.

Conclusion to Part III

We have seen that failure to meet RHNA goals is not due to anything the cities and counties are, or are not doing but simply because RHNA is absurdly unrealistic in it's projections for growth.  RHNA needs to be radically altered, or simply abolished as it is extraordinarily unrealistic.